ai/prompt-io/claude/20260813T001102Z_27c34aeb_prompt_io.md

@@ -3,8 +3,8 @@ model: claude-opus-5
 service: claude
 session: 7b9c97c4-fff7-4ac4-97fb-35720453308e
 timestamp: 2026-08-13T00:11:02Z
-git_ref: 27c34aebb615c30d4039fa399f4ce2766ed7ba2c
+git_ref: 27c34aeb
-scope: code
+scope: docs+code
 substantive: true
 raw_file: 20260813T001102Z_27c34aeb_prompt_io.raw.md
 ---

ai/prompt-io/claude/20260813T001102Z_27c34aeb_prompt_io.raw.md

@@ -2,15 +2,15 @@
 model: claude-opus-5
 service: claude
 timestamp: 2026-08-13T00:11:02Z
-git_ref: 27c34aebb615c30d4039fa399f4ce2766ed7ba2c
+git_ref: 27c34aeb
-diff_cmd: git diff 3ad7e7e5dc28aa336499725a03744fc10c6c0026..27c34aebb615c30d4039fa399f4ce2766ed7ba2c
+diff_cmd: git diff main..ng_tpts_planning
 ---
 
 # Raw output — next-gen tpt-backend implementation plans
 
 ## Generated planning docs
 
-> `git diff 3ad7e7e5dc28aa336499725a03744fc10c6c0026..27c34aebb615c30d4039fa399f4ce2766ed7ba2c -- ai/tpt-backends/`
+> `git diff main..ng_tpts_planning -- ai/tpt-backends/`
 
 Five markdown docs. `00_shared_backend_contract.md` is normative
 and the other three are written against it so they can be worked
@@ -43,7 +43,7 @@ independently:
 
 ## Generated example code
 
-> `git diff 3ad7e7e5dc28aa336499725a03744fc10c6c0026..27c34aebb615c30d4039fa399f4ce2766ed7ba2c -- examples/multihost/wg_lan/`
+> `git diff main..ng_tpts_planning -- examples/multihost/wg_lan/`
 
 - `wg_maddr.py` — `WGTunnelledAddr(msgspec.Struct, frozen=True)`
   carrying `bearer: tuple[str, int]`, `peer_pubkey: str`,

ai/prompt-io/opencode/20260831T022317Z_768b5316_prompt_io.md

@@ -1,72 +0,0 @@
----
-model: gpt-5.6-sol
-service: opencode
-session: f3a5eef0-9695-48ab-a336-861187473f8a
-timestamp: 2026-08-31T02:23:17Z
-git_ref: ng_tpts_planning
-diff_cmd: git diff 768b531662a88037e3988885102d2ff24a7ce34e..HEAD
-scope: code
-substantive: true
-raw_file: 20260831T022317Z_768b5316_prompt_io.raw.md
----
-
-## Prompt
-
-The human asked to continue the retained PR landing sequence and
-selected "Repair all blockers" after a read-only review of PR #492.
-They then corrected the TIPC assessment:
-
-> oh one thing on the tipc findings, they may have already been
-> addressed in the downstream patch branch no?
-
-The repair therefore had to use downstream #493 as implementation
-evidence, avoid duplicating solved TIPC work, preserve the dirty
-downstream worktree, and repair the remaining shared, QUIC,
-WireGuard, lock, and provenance defects without committing,
-pushing, or publishing.
-
-## Response summary
-
-Reworked PR #492's plans and executable WireGuard example into a
-consistent landing candidate. The shared and TIPC documents now
-match current runtime/downstream behavior while naming remaining
-risks. The QUIC plan now defines transport ownership across the
-whole actor lifecycle. The WireGuard example validates keys and
-binds both hosts and the service actor correctly without elevating
-Tractor. The dependency lock and historical provenance pointers
-are synchronized.
-
-## Files changed
-
-- `ai/tpt-backends/00_shared_backend_contract.md` - current backend
-  contract, dispatch, capability, and locality rules.
-- `ai/tpt-backends/01_tipc_backend.md` - #493 reconciliation and
-  bounded unresolved work.
-- `ai/tpt-backends/02_quic_iroh_backend.md` - lifecycle-safe QUIC
-  and UniFFI implementation plan.
-- `ai/tpt-backends/03_wg_tunnel_bindspace.md` - corrected address
-  registry and privilege semantics.
-- `ai/tpt-backends/README.md` - refreshed plan summaries.
-- `examples/multihost/wg_lan/README.md` - unprivileged two-host
-  setup and accurate verification limits.
-- `examples/multihost/wg_lan/host_a_srv.py` - local key check,
-  overlay child bind, and stable RPC exposure.
-- `examples/multihost/wg_lan/host_b_client.py` - peer key check,
-  host-B bind, and explicit missing-service failure.
-- `examples/multihost/wg_lan/wg_maddr.py` - strict parsing and
-  asynchronous role-specific key inspection.
-- `uv.lock` - exact `py-multiaddr` Git source resolution.
-- `ai/prompt-io/claude/20260813T001102Z_27c34aeb_prompt_io.md` -
-  valid scope and immutable Git reference.
-- `ai/prompt-io/claude/20260813T001102Z_27c34aeb_prompt_io.raw.md`
-  - immutable historical diff pointers.
-
-## Human edits
-
-The human chose the full repair path rather than reducing the PR
-to planning documents or publishing the initial review. They also
-identified that the initial TIPC review had not accounted for the
-downstream implementation branch. That correction materially
-changed the work: solved TIPC items were backported into the plan,
-remaining defects were separated from implemented behavior, and
-the dirty downstream worktree was kept read-only.

ai/prompt-io/opencode/20260831T022317Z_768b5316_prompt_io.raw.md

@@ -1,60 +0,0 @@
----
-model: gpt-5.6-sol
-service: opencode
-timestamp: 2026-08-31T02:23:17Z
-git_ref: ng_tpts_planning
-diff_cmd: git diff 768b531662a88037e3988885102d2ff24a7ce34e..HEAD
----
-
-# Raw output - repair PR #492 for landing
-
-## Generated changes
-
-> `git diff 768b531662a88037e3988885102d2ff24a7ce34e..HEAD -- ai/tpt-backends/`
-
-- Reconciled the shared backend contract with current address,
-  dispatch, capability, locality, and listener-rebind APIs.
-- Updated the TIPC plan from downstream #493 implementation and
-  tests, preserving unresolved registrar election, collision,
-  locality, and socket-cleanup work as explicit follow-ups.
-- Reworked the QUIC plan around a launch-time transport bootstrap,
-  one actor-owned endpoint, a transport nursery spanning parent
-  dial through deregistration, supervised UniFFI cleanup, complete
-  routable addresses, connection leases, and listener-owned tasks.
-- Corrected the WireGuard bindspace plan to distinguish the
-  build-registered address registry from runtime capability.
-
-> `git diff 768b531662a88037e3988885102d2ff24a7ce34e..HEAD -- examples/multihost/wg_lan/`
-
-- Hardened WireGuard key conversion and parsing with strict
-  32-byte validation, lazy protocol lookup, and explicit rejection
-  of unsupported nested tunnel descriptors.
-- Made interface inspection asynchronous, bounded at its
-  cancellation request, role-specific, and separable from
-  privileged preflight commands.
-- Corrected host-local binds, subactor overlay publication, stable
-  RPC module exposure, and missing-actor handling in the two-host
-  example.
-- Updated the README so Tractor remains unprivileged and the local
-  versus remote overlay roles are explicit.
-
-> `git diff 768b531662a88037e3988885102d2ff24a7ce34e..HEAD -- pyproject.toml uv.lock ai/prompt-io/`
-
-- Regenerated `uv.lock` for the exact merged `py-multiaddr` WireGuard
-  codec revision.
-- Replaced mutable historical prompt diff pointers with immutable
-  refs and normalized the substantive scope without rewriting the
-  historical raw response.
-
-## Verification output
-
-- `git diff --check`: passed.
-- `uv lock --check`: passed.
-- Ruff on the WireGuard example directory: passed.
-- Python compilation of the WireGuard example directory: passed.
-- Tractor imported from the PR worktree's existing environment.
-- WireGuard address round-trip, local/peer role checks, malformed
-  base64 rejection, and nested-tunnel rejection: passed.
-- Three independent final re-reviews reported no actionable
-  findings in the shared/TIPC, QUIC, or WireGuard slices.
-- No live TIPC, iroh/UniFFI, or WireGuard network test was run.

ai/tpt-backends/00_shared_backend_contract.md

@@ -33,35 +33,19 @@ doc in the same PR.
 
 ## 1. The backend duck-type (empirical, from `_tcp.py`/`_uds.py`)
 
-A transport backend is **one module** under `tractor/ipc/`.
+A transport backend is **one module** under `tractor/ipc/`
-There is no ABC to subclass and no plugin entrypoint; wiring is by
+exposing exactly four things. There is no ABC to subclass and no
-explicit table registration (§2) plus one piece of reflection
+plugin entrypoint; wiring is by explicit table registration
-(§1.3).
+(§2) plus one piece of reflection (§1.3).
-
-Keep two contracts distinct:
-
-- `tractor.discovery._addr.Address` is a static `Protocol`. It
-  declares address-wrapper members including `namespace`,
-  `open_listener()` and `close_listener()`.
-- the runtime's empirical contract is what `_tcp.py`, `_uds.py`
-  and `_server.py` actually call. The current address classes do
-  not implement every declared `Address` member: listener
-  lifecycle is module-level, `def_bindspace` is used despite not
-  being declared by the `Protocol`, and `namespace` remains
-  aspirational.
-
-Until those surfaces are deliberately reconciled, implement the
-empirical module contract below and update the static `Protocol`
-only when the runtime really consumes the new member. Do not claim
-that structural conformance alone defines a backend.
 
 ### 1.1 `class <Proto>Address(msgspec.Struct, frozen=True)`
 
-The runtime-consumed address-wrapper surface is:
+Structurally conforms to the `Address` `Protocol` in
+`tractor/discovery/_addr.py:82`. Required surface:
 
 | member | kind | notes |
 | --- | --- | --- |
-| `proto_key` | `ClassVar[str]` | internal transport key, e.g. `'tcp'`, `'uds'` |
+| `proto_key` | `ClassVar[str]` | the wire/registry key, e.g. `'tcp'`, `'uds'` |
 | `unwrapped_type` | `ClassVar[type]` | the primitive tuple shape |
 | `def_bindspace` | `ClassVar` | default bindspace value |
 | `is_valid` | `@property -> bool` | "is this a *dialable/bindable* addr" |
@@ -97,32 +81,22 @@ Hard constraints learned from the existing two:
   paper over it at best.
 
   **The fix, and the recommended prerequisite for all three
-  backends: make the unwrapped form carry an explicit internal
+  backends: make the unwrapped form carry an explicit
-  proto-key** — `('tcp', host, port)`,
+  proto-key, using the `multiaddr` protocol name as the
-  `('uds', filedir, filename)`, `('tipc', stype, inst, scope)`.
+  canonical spelling** — `('tcp', host, port)`,
-  The tag must be a `TransportProtocolKey`/registry key. In
+  `('unix', path)`, `('udp', ...)`, `('tipc', stype, inst,
-  particular it is **`'uds'`, not the external multiaddr spelling
+  scope)`. Then `wrap_address()` collapses from an
-  `'unix'`**. If a wire or display format uses a different name,
+  order-sensitive `match` to `_address_types[addr[0]]`, and the
-  name that translation explicitly; today `_multiaddr.py` maps
+  whole collision class stops existing. Note this *also* aligns
-  internal `uds` to external `/unix/`. Then `wrap_address()` can
+  the on-wire form with `mk_maddr()`/`parse_maddr()`, so the two
-  dispatch through `_address_types[addr[0]]` without an
+  representations stop being independent inventions.
-  order-sensitive shape match.
 
   Two consequences to plan for:
-  - it's a **wire-format change**. Widen and keep synchronized
+  - it's a **wire-format change** (`SpawnSpec`,
-    `discovery._addr.UnwrappedAddress` and the duplicate wire
+    `_root_mailbox`, `_registry_addrs`) plus every test fixture
-    alias in `msg.types`; change `SpawnSpec.reg_addrs` and
+    and downstream config (`piker`'s `[network]` table). It
-    `.bind_addrs`, not only `_root_mailbox` and
+    wants its **own migration commit, landed before any new
-    `_registry_addrs`. Audit the related `RuntimeVars`
+    backend**, not smuggled into one.
-    `_root_mailbox`/`_root_addrs` annotations, `Actor.reg_addrs`
-    and accept-address annotations, channel/spawn signatures,
-    fixtures, and downstream config (`piker`'s `[network]`
-    table). `msgspec` rejects a union containing multiple
-    array-like tuple shapes, so #493 used
-    `tuple[str|int, ...]` as the truthful transitional wire type;
-    the complete proto-key migration can restore per-proto
-    validation. This wants its **own migration commit, landed
-    before any new backend**, not smuggled into one.
   - it's the moment to **stop handing raw unwrapped tuples to
     users at all.** The long-term shape is: `Address` subtypes
     are the public currency and `UnwrappedAddress` becomes an
@@ -130,8 +104,8 @@ Hard constraints learned from the existing two:
     `ipaddress` uses (you pass `IPv4Address`, not a 4-tuple).
     Public API should accept `Address|maddr-str` and treat bare
     tuples as legacy-tolerated input, ideally deprecated.
-- **`.get_random()` must not deterministically alias without a
+- **`.get_random()` must be collision-free without a live
-  live runtime.** See the `UDSAddress.get_random()` uuid-token
+  runtime.** See the `UDSAddress.get_random()` uuid-token
   comment (`_uds.py:207-220`): with no `current_actor()` the
   sockname degenerates to a pure fn of `(prefix, pid)` and two
   calls in one proc alias. Mix in a `uuid4().hex[:8]` token.
@@ -188,11 +162,10 @@ if (unwrapped := lstnr.socket.getsockname()) != self.addr.unwrap():
 ```
 
 i.e. it assumes `lstnr.socket.getsockname()` exists and that its
-return value is a valid `from_addr()` input. That is false for
+return value is a valid `from_addr()` input. This is fine for
-TIPC, whose listener sockname is an undialable port ID, and for
+TIPC (§3 of plan 01) and **is the main integration hazard for
-non-socket iroh. Both plans must use the explicit backend rebind
+iroh** (§3 of plan 02) — plans that break it must say so
-policy added at this integration point rather than pretending a
+explicitly and propose the upstream `_server.py` patch.
-sockname is always an address replacement.
 
 ### 1.4 `class Msgpack<Proto>Stream(MsgpackTransport)`
 
@@ -251,26 +224,20 @@ path.** This is why plan 01 is small and plan 02 is not.
 
 ---
 
-## 2. Registration and policy wiring
+## 2. Registration tables (the full wiring checklist)
 
-Adding a backend requires this complete audit. Not every item
+Adding a backend touches these and only these:
-changes for every backend, but none may be assumed from the others:
 
 1. `tractor/runtime/_state.py:46`
    `TransportProtocolKey = Literal['tcp', 'uds', ...]` — add the
-   internal key. This `Literal` is the **declared protocol-key
+   key. This `Literal` is the canonical set; `_testing/pytest.py`
-   set**, not proof that a backend is usable on this host.
+   drives `--tpt-proto` validation off `_addr._address_types`,
-2. `tractor/discovery/_addr.py` `_address_protos` and
+   and the spawn-backend fixture already models the
-   `_address_types: dict[str, Type[Address]]` — register
+   "drive-the-set-from-the-Literal" pattern
-   `'<key>': <Proto>Address`. `_address_types` is a plain
+   (`pytest.py:870-880`) — do the same rather than hardcoding.
-   **`dict`, not a `bidict`**, and represents the backends this
+2. `tractor/discovery/_addr.py:173` `_address_types: bidict` —
-   build registers for import and dispatch. UDS is conditional on
+   `{'<key>': <Proto>Address}`. Note it is a **`bidict`**, so
-   `HAS_UDS`, while TIPC can remain registered on a host where its
+   the mapping must stay 1:1.
-   kernel support is unavailable. An importable backend with a
-   runtime capability requirement therefore needs a separate
-   availability check. Never conflate this dispatch registry with
-   either host usability or the declared `TransportProtocolKey`
-   universe.
 3. `tractor/discovery/_addr.py:181` `_default_lo_addrs` —
    `'<key>': <Proto>Address.get_root().unwrap()`.
    ⚠️ this dict is built at **import time**, so
@@ -283,7 +250,7 @@ changes for every backend, but none may be assumed from the others:
    add a case iff your `unwrapped_type` isn't already uniquely
    matched. **Preferably do the proto-key migration in §1.1
    first**, after which this step becomes a one-line
-    `_address_types` lookup instead of an order-sensitive `case`.
+   `_address_types` entry instead of an order-sensitive `case`.
 5. `tractor/ipc/_types.py` — `Address` union alias,
    `_msg_transports` list, `_key_to_transport[('msgpack', key)]`,
    `_addr_to_transport[<Proto>Address]`.
@@ -295,17 +262,9 @@ changes for every backend, but none may be assumed from the others:
    `parse_maddr()`.
 8. `tractor/ipc/__init__.py` — re-export if the backend has a
    public surface.
-9. `tractor/discovery/_api.py::_is_local_addr()` and
+9. `tractor/_testing/addr.py::get_rando_addr()` — per-proto
-   `prefer_addr()` — define and test the backend's locality and
+   branch so the whole suite can run under `--tpt-proto <key>`.
-   selection tier. The current order is UDS, local TCP, then
+10. `pyproject.toml` — new deps go in an **optional extra**, never
-   remote. A new backend must not silently fall into `remote` by
-   accident: for example TIPC node scope is local, cluster scope
-   is not known-local, and an observed address with unknown scope
-   must not be promoted. Preserve the last-registered tie-break
-   unless intentionally changing policy.
-10. `tractor/_testing/addr.py::get_rando_addr()` — per-proto
-    branch so the whole suite can run under `--tpt-proto <key>`.
-11. `pyproject.toml` — new deps go in an **optional extra**, never
     in `[project].dependencies`. See §5.
 
 ## 3. Where the `trio.SocketListener` assumption is load-bearing
@@ -380,10 +339,7 @@ dep-free, or make that table lazy.
   `_state._def_tpt_proto` + `_runtime_vars['_enable_tpts']`
   (`pytest.py:807-835`). Adding the key to `_address_types` is
   what makes `--tpt-proto <key>` legal (`pytest.py:795-800`
-  asserts the lookup). Thus CLI acceptance follows the
+  asserts the lookup).
-   build-registered `_address_types`, while type-level declarations
-   follow `TransportProtocolKey` and host usability follows each
-   backend's capability probe; test all three layers separately.
 - The **acceptance bar** for every backend is: the *entire*
   existing suite passes under `--tpt-proto <key>`, unmodified.
   That is the whole point of the abstraction. Backend-specific
@@ -397,11 +353,8 @@ dep-free, or make that table lazy.
   `OSError(97, 'Address family not supported by protocol')`
   because the `tipc` module isn't loaded. Put the predicate in
   the backend module (so apps can use it too), not in the test.
-- New pytest marks must be registered in
+- New pytest marks must be registered in `pyproject.toml`, per
-  `_testing/pytest.py::pytest_configure()` with
+  the project's fix-warnings-at-source rule (gh #469).
-  `config.addinivalue_line()`, alongside the existing custom
-  marks. The repo has no `pyproject.toml` marker table. This is
-  still part of the fix-warnings-at-source rule (gh #469).
 
 ## 7. Code style (non-negotiable, matches the repo)
 

ai/tpt-backends/01_tipc_backend.md

@@ -4,18 +4,13 @@ Tracks gh [#378]. Prereq reading:
 [`00_shared_backend_contract.md`](./00_shared_backend_contract.md).
 
 **Thesis**: TIPC is the *cheapest* new backend we can add and
-gives us kernel-native service-name publication, known-address
+simultaneously the only one that gives us cluster-wide service
-dialling, and topology events. Those are primitives for reducing
+discovery **for free, in the kernel**, replacing (for
-registrar traffic; they do **not** by themselves replace
+TIPC-capable deployments) the whole `tractor.discovery`
-`tractor.discovery`, derive an actor's address from its name, or
+registrar round-trip with a `bind()`/`connect()` on a
-elect one registrar. It is stdlib-only: zero new dependencies.
+*service name*. It is stdlib-only: zero new dependencies.
-
-This plan is reconciled against downstream PR [#493]'s code and
-tests. Treat that implementation as prior art without mistaking
-implemented transport primitives for completed discovery policy.
 
 [#378]: https://github.com/goodboy/tractor/issues/378
-[#493]: https://github.com/goodboy/tractor/pull/493
 
 ---
 
@@ -79,12 +74,10 @@ The design decision that makes this backend coherent:
 > ever an *observed* address (`getpeername()`), never a
 > user-facing one.**
 
-This is the "leverage the built-in discovery machinery" part of
+This is exactly the "leverage the built-in discovery machinery"
-#378: publishing a bind is kernel name-table registration and
+ask in #378: publishing a bind *is* registration, and
-`connect()` on an already-known name is a kernel lookup, with no
+`connect()` on a name *is* a lookup, with no registrar actor in
-registrar actor on that **dial** path. Mapping an application name
+the loop.
-to that address and maintaining Tractor's actor registry remain
-separate work (§5).
 
 ### 2.2 the struct
 
@@ -96,12 +89,12 @@ class TIPCAddress(
     _stype: int                    # TIPC "type" == service class
     _instance: int                 # service instance within the type
     _scope: int = TIPC_CLUSTER_SCOPE
-    # observed-only, excluded from the unwrapped service identity
+    # observed-only, never part of identity/equality-by-intent
     maybe_node: int|None = None    # from TIPC_ADDR_ID getpeername()
     maybe_ref: int|None = None
 
     proto_key: ClassVar[str] = 'tipc'
-    unwrapped_type: ClassVar[type] = tuple[str, int, int, int]
+    unwrapped_type: ClassVar[type] = tuple[str, int]
     def_bindspace: ClassVar[int] = TIPC_CLUSTER_SCOPE
 ```
 
@@ -113,8 +106,8 @@ shape as `TCPAddress`*, so `wrap_address()`'s
 `case (str(), int())` steals it. This backend is therefore the
 forcing function for the contract-doc's conclusion (§1.1):
 
-> **make the unwrapped form carry the explicit internal
+> **make the unwrapped form carry an explicit proto-key, spelled
-> `TransportProtocolKey`.**
+> with the `multiaddr` protocol name.**
 
 ```python
 def unwrap(self) -> tuple[str, int, int, int]:
@@ -122,31 +115,11 @@ def unwrap(self) -> tuple[str, int, int, int]:
 ```
 
 `wrap_address()` then dispatches `_address_types[addr[0]]` and
-the collision class disappears. The complete all-backend change
+the collision class disappears. **This is a prerequisite
-is a prerequisite migration; #493 necessarily carried the
+migration commit, not part of this backend** — see contract §1.1
-transitional `UnwrappedAddress`/`SpawnSpec.reg_addrs`/
+for its blast radius (wire format + every fixture + `piker`
-`.bind_addrs` widening needed for TIPC. See contract §1.1 for the
+config) and for the follow-on "stop handing raw tuples to users
-remaining runtime annotations, fixtures and `piker` config. Here
+at all, à la `ipaddress`" direction.
-`tipc` is both the internal and external spelling; UDS remains
-internally `uds` and translates explicitly to external `/unix/`.
-
-`msgpack` decodes tuples as lists, so both forms are part of the
-round-trip contract. Match only the exact three- or four-element
-tagged shapes and test all four routes:
-
-```python
-case (
-    ('tipc', int() as stype, int() as inst, int() as scope)
-    |
-    ['tipc', int() as stype, int() as inst, int() as scope]
-):
-    ...
-```
-
-Also test the scope-defaulted three-element form through
-`TIPCAddress.from_addr()`, and tuple/list forms through the global
-`wrap_address()`. A normal two-element TCP/UDS address whose first
-element happens to be `'tipc'` must retain its classic dispatch.
 
 ⚠️ an earlier revision of this plan proposed a self-tagging
 `('tipc:<stype>:<scope>', instance)` string-prefix hack with an
@@ -177,28 +150,25 @@ treatment (`_uds.py:242`).
 - `_instance` for `get_random()`: TIPC gives us no
   kernel-assigned-instance analogue of `port=0`, so we must
   choose. Use a *pure* fn of the actor identity so it is
-  reproducible and well-distributed, **not collision-free**:
+  reproducible and collision-free:
   ```python
-  # 32-bit instance derived from the actor's Aid.uid, or from a
+  # 32-bit instance derived from the actor's uuid4 (+ pid when
-  # per-call token + pid when there is no live runtime.
+  # there's no live runtime, per the UDS precedent).
   inst: int = int.from_bytes(
       blake2b(seed.encode(), digest_size=4).digest(),
       'big',
   )
   ```
-  where `seed = '.'.join(actor.aid.uid)` if
+  where `seed = f'{actor.aid.name}@{pid}'` if
   `current_actor(err_on_no_runtime=False)` else
   `f'{prefix}.{uuid4().hex[:8]}@{pid}'`. Must avoid the reserved
   low range: `inst = 64 + (inst % (2**32 - 64))`.
-  The UUID is load-bearing because TIPC names are cluster-wide
-  while PIDs are only host-local: `(actor name, pid)` can alias on
-  different hosts.
   ⚠️ *unlike* `port=0`, a collision here surfaces as a
   successful-but-shared publication (TIPC allows multiple
   binders on the same name and round-robins!) rather than
   `EADDRINUSE`. That is a silent-crosstalk failure mode; §7 has
-  a statistical test and §9 records the unresolved recovery work
+  the test that proves the 4-byte digest is enough and §9 has
-  in [#501].
+  the mitigation if it isn't.
 - `_scope`: `TIPC_NODE_SCOPE` for a same-host-only actor (the
   UDS-equivalent), `TIPC_CLUSTER_SCOPE` (default) for
   cluster-visible. **This is `.bindspace`**:
@@ -219,9 +189,7 @@ treatment (`_uds.py:242`).
 @property
 def is_valid(self) -> bool:
     return (
-        self._instance > 0
+        self._instance != 0
-        and
-        self._stype > 0
         and
         self._stype not in _tipc_reserved_stypes  # {0, 1, ...}
         and
@@ -268,11 +236,10 @@ Notes / hazards:
 - **no `close_listener()` needed** — nothing to unlink. Omit the
   function entirely (contract §1.2: absence means implicit).
   Withdrawal of the published name happens on socket close.
-- `SocketListener.__init__` calls
+- ⚠️ `SocketListener.__init__` will try
-  `getsockopt(SOL_SOCKET, SO_ACCEPTCONN)`. The live-kernel probe
+  `getsockopt(SOL_SOCKET, SO_ACCEPTCONN)`. If TIPC rejects it,
-  used by #493 answers `1`; retain the unit test so a kernel-side
+  trio's `except OSError: pass` covers us. Assert this in a
-  change is visible rather than relying on trio's suppressed-
+  unit test rather than assuming.
-  `OSError` carve-out.
 - Wrap the bind in a `_reraise_as_connerr()`-style `@cm` (copy
   the `_uds.py:256` pattern) so `EADDRINUSE`-ish and
   `EAFNOSUPPORT` become `ConnectionError` with the addr in the
@@ -291,31 +258,33 @@ returns a `TIPC_ADDR_ID`-flavoured 5-tuple (the port id), *not*
 the name-seq we bound. So the `!=` is **always true** and
 `from_addr()` will be handed a 5-tuple.
 
-`TIPCAddress.from_addr()` must accept only proto-keyed service
+Handle it inside `TIPCAddress.from_addr()` — do **not** patch
-names. It must reject a bare port ID because no conversion can
+`_server.py`:
-recover `(stype, instance)`:
 
 ```python
 @classmethod
 def from_addr(cls, addr) -> TIPCAddress:
     match addr:
-        # our proto-keyed tuple or decoded-list wire form
+        # our own unwrapped form
-        case (
+        case (str() as tag, int() as inst) if tag.startswith('tipc:'):
-            ('tipc', int() as stype, int() as inst, int() as scope)
+            _, stype, scope = tag.split(':')
-            |
+            return TIPCAddress(int(stype), inst, int(scope))
-            ['tipc', int() as stype, int() as inst, int() as scope]
-        ):
-            return TIPCAddress(stype, inst, _norm_scope(scope))
 
-        # a bare kernel-observed TIPC_ADDR_ID 5-tuple has no
+        # a kernel-observed TIPC_ADDR_ID 5-tuple: keep the
-        # service identity to annotate.
+        # *service* identity we already know and only annotate
+        # the observed port-id.
         case (int() as atype, *rest) if atype == socket.TIPC_ADDR_ID:
-            raise ValueError(...)
+            ...
 ```
 
-The `TIPC_ADDR_ID` case cannot reconstruct `(stype, instance)`.
+The `TIPC_ADDR_ID` case cannot reconstruct `(stype, instance)`
-The resolution is the explicit listener-rebind policy added ahead
+— that info isn't in a port id. So `from_addr()` alone is
-of the backend in #493:
+insufficient for the reconciliation path. **Resolution**: make
+`from_addr()` raise a clear `ValueError` for the bare
+`TIPC_ADDR_ID` case, and instead prevent the reconciliation
+from firing by having `start_listener()` return a listener
+whose `getsockname()` we never need — i.e. land this two-line
+upstream fix in `_server.py:664`:
 
 ```python
 if (
@@ -331,13 +300,12 @@ behaviour exactly). Rationale: the reconciliation exists *only*
 to learn the kernel-assigned port for `port=0` TCP binds (its
 own comment says so, `_server.py:662`); TIPC has no such
 late-binding, so opting out is semantically right rather than a
-hack. Keep the guard test that TCP's `port=0` behaviour is
+hack. **Land this as its own commit, ahead of the backend**,
-unchanged.
+with a test that `tcp`'s `port=0` behaviour is unchanged.
 
-Do **not** annotate `Endpoint.addr` from `getsockname()`: the
+Keep the observed port-id available anyway: annotate
-listener endpoint must remain the dialable service name. Port IDs
+`ep.addr = ep.addr.with_port_id(*getsockname()[1:3])` (a pure
-are observed only on connected streams and may annotate a copy via
+`msgspec.structs.replace()` helper) purely for logging/repr.
-`with_port_id()` purely for logging/repr.
 
 ### 3.3 `MsgpackTIPCStream`
 
@@ -372,12 +340,11 @@ class MsgpackTIPCStream(MsgpackTransport):
                 0,              # domain: 0 == "anywhere in scope"
                 destaddr._scope,
             ))
-            stream = trio.SocketStream(sock)
+        return cls(
-            return cls(
+            trio.SocketStream(sock),
-                stream,
+            prefix_size=prefix_size,
-                prefix_size=prefix_size,
+            codec=codec,
-                codec=codec,
+        )
-            )
 ```
 
 - reuse `trio._highlevel_open_unix_stream.close_on_error` (the
@@ -396,11 +363,11 @@ class MsgpackTIPCStream(MsgpackTransport):
     leave at default, we have `trio` cancel scopes.
   - `TIPC_DEST_DROPPABLE = 0` on the connection so undeliverable
     msgs come back as errors rather than being silently dropped.
-- **`connect_to()` on a name with no publisher**: the live-kernel
+- **`connect_to()` on a name with no publisher**: TIPC returns
-  result is immediate `EHOSTUNREACH`. Python exposes that as a
+  `ECONNREFUSED`/`EHOSTUNREACH` promptly (no SYN-timeout wait),
-  bare `OSError`, not a `ConnectionError` subtype, so
+  which is *better* discovery-ping behaviour than TCP. Confirm
-  `_reraise_as_connerr()` is load-bearing for contract §4. Keep
+  the errno and make sure it surfaces as `ConnectionError`
-  the exact errno and normalization under test.
+  (contract §4 — the registrar ping path depends on it).
 
 ### 3.4 `get_stream_addrs()`
 
@@ -418,27 +385,29 @@ Problem: neither end's port-id tells us the *service name*. The
 `laddr`/`raddr` are used for logging, `Channel.raddr`,
 `Server._peers` keying-adjacent repr, and `maddr`. Design:
 
-- `get_stream_addrs()` converts both socket results into
+- the **connecting** side knows the destaddr it dialled →
-  **observed-only** addresses: `_stype`/`_instance` use the
+  `connect_to()` overrides `_raddr` after construction with the
-  `TIPC_NAME_UNKNOWN = -1` sentinel and `maybe_node`/`maybe_ref`
+  known-good `TIPCAddress`, exactly as
-  carry the port ID. Such addresses are invalid for dialling.
+  `MsgpackUDSStream.connect_to()` does for the peer-pid case
-- the **connecting** side knows the service name it dialled, so
+  (`_uds.py:539-543`).
-  `connect_to()` replaces `_raddr` after construction with that
+- the **accepting** side does not know the peer's service name
-  known `TIPCAddress` while retaining the constructor's one
+  from the socket. Two honest options:
-  tolerant port-ID observation. Do not call `getpeername()` a
+  - **(a) accept it: `raddr` carries only `(node, ref)`** via
-  second time: the peer can withdraw between the two calls.
+    `maybe_node`/`maybe_ref`, `_stype/_instance` set to a
-- the **accepting** side genuinely cannot recover the peer's
+    sentinel `-1`, and `__repr__` renders
-  service name from a port ID. Keep the observed-only `raddr`;
+    `TIPCAddress[<peer-node:0x...>:<ref>]`. The `Aid` from the
-  the handshake's `Aid` supplies logical identity. Piggybacking a
+    handshake already gives us the peer's logical identity, so
-  bound name in the handshake is outside this backend.
+    nothing in the runtime actually *needs* the peer's service
-- `laddr` is observed-only as well. It is used for repr/logging,
+    name. **Recommended.**
-  not to replace the endpoint's known service name.
+  - (b) piggyback the peer's own bound name in the handshake.
-- unlike TCP/UDS, TIPC can answer `ENOTCONN` from
+    Rejected for this PR: touches `Aid`/msg-spec.
-  `getpeername()` after a connect-then-drop. This lookup happens
+- `laddr` on the accepting side: the `Endpoint` knows its own
-  during `MsgpackTransport` construction, before handshake error
+  `addr`; but `get_stream_addrs()` is a `@classmethod` with only
-  tolerance. Wrap `getsockname()` and `getpeername()` in a
+  the stream. Use `TIPC_ADDR_ID` for `laddr` too and let
-  tolerant helper and degrade to a port-ID-less observed address;
+  `Endpoint.peer_tpts` keying (which is by *peer* addr) still
-  a dropped peer must cost an observation, not kill the actor.
+  work. Verify nothing asserts `laddr == ep.addr` — grep for
+  `.laddr` uses before committing (`_server.py`'s
+  `con_status` logging, `Channel.pformat()`).
 
 ---
 
@@ -475,47 +444,43 @@ the maddr stays 2-segment like `/unix/...`.
 
 ---
 
-## 5. Discovery primitives and explicit limits
+## 5. Discovery: the actually-interesting part
 
-The backend provides independently-shippable kernel primitives.
+Two independently-shippable layers. **Layer A is in scope for
-Neither primitive alone implements Tractor's actor-name discovery,
+the first PR; layer B is a fast-follow.**
-registry ownership, or registrar election.
 
 ### 5.1 Layer A — "discovery by bind" (free)
 
 Because `bind(TIPC_ADDR_NAMESEQ)` publishes and
-`connect(TIPC_ADDR_NAME)` resolves, a caller that **already knows**
+`connect(TIPC_ADDR_NAME)` resolves, a `tractor` tree whose
-a TIPC service address can dial it without a registrar lookup.
+`registry_addrs` are TIPC service names needs **no registrar
-This is narrower than registrar-less `find_actor(name)`:
+liveness at all** for the connect path: `find_actor()`'s
+"connect to the registrar and ask" becomes "connect to the
+service name directly". Concretely:
 
-- `tractor.discovery._api.find_actor()` and peers still query a
+- `tractor.discovery._api.find_actor()` etc. keep working
-  registrar; #493 does not change them.
+  unchanged (they go through the registrar), *and*
-- deriving a stable service address from `(name, uuid)` and
+- a new, TIPC-only fast path becomes possible: derive an actor's
-  dialling it directly is follow-up [#499]. The mapping must be
+  service name from its `(name, uuid)` and dial it without any
-  documented and cross-language stable.
+  registrar hop.
-- `registry_addrs` still identify registrars. Connecting to a
-  known registrar by TIPC name removes no registrar bookkeeping
-  or ownership semantics.
 
-There is also an unresolved **split-brain election** problem.
+Do **not** build the fast path in PR 1. Instead, prove the
-Duplicate TIPC name publication succeeds and round-robins, so two
+property with a test (§7.4) and file the follow-up: it changes
-roots can both probe an unoccupied registrar name, both bind it,
+`discovery` semantics (name→instance derivation must be a
-and both believe they won. The backend provides no atomic
+documented, stable, cross-language-able hash) and deserves its
-compare-and-publish, lease, quorum, or deterministic winner. A
+own design.
-topology subscription can reveal multiple publisher port IDs but
-does not elect or fence one. Do not describe registrar election as
-solved until a separate protocol closes this race.
 
 ### 5.2 Layer B — the topology service (`TIPC_TOP_SRV`)
 
-This is the push primitive behind #378's "end game cluster proto"
+This is what makes #378's "end game cluster proto" claim real:
-direction: a subscription to kernel name-table publish/withdraw
+a *subscription* to name-table events, i.e. push-based
-events. #493 implements `open_topology_events()`; consuming that
+`register`/`deregister` for free, replacing the registrar's
-feed in `discovery._registry` is follow-up [#496]. Until then it
+polled `find_actor()`.
-does not replace registrar state or `find_actor()`.
 
-Mechanics, verified against `linux/include/uapi/linux/tipc.h`,
+Mechanics (verify each field against
-`net/tipc/topsrv.c` and #493's live-kernel probe:
+`linux/include/uapi/linux/tipc.h` + `net/tipc/topsrv.c` at
+implementation time — the struct layout below is from the uapi
+header and the byte-order caveat is real):
 
 ```python
 # SOCK_SEQPACKET connected to the topology server
@@ -533,25 +498,22 @@ await sock.connect((
 #   __u32 filter;              /* TIPC_SUB_{PORTS,SERVICE,CANCEL} */
 #   char  usr_handle[8];
 # }                            /* == 28 bytes */
-_SUBSCR_FMT: str = '=5I8s'
+_SUBSCR_FMT: str = '=IIIII8s'   # ⚠ 5*I is 20 -> use '=5I8s'
 ```
 
-- **byte order**: #493's live-kernel probe verified native
+- **byte order**: the topology server historically accepts both
-  standard-size (`'='`) packing for publish and withdraw events.
+  host and swapped order and auto-detects; modern kernels are
-  Use `'=5I8s'` for the 28-byte subscription. Do not retain the
+  strict-ish. Pack native (`'='`) first, and if the server
-  speculative `'>'` retry/probe as if it were required. Preserve
+  closes the connection immediately, retry with `'>'`. Encode
-  the earlier `# ?TODO` to verify the deterministic rule directly
+  that as a one-time probe helper
-  against `net/tipc/topsrv.c`; it is source-audit work, not a
+  `_detect_topsrv_endianness()` cached at module level — and
-  runtime retry requirement.
+  put a `# ?TODO` pointing at `net/tipc/topsrv.c` for someone
+  to make it deterministic.
 - **events**: `struct tipc_event` is `event: u32`,
   `found_lower: u32`, `found_upper: u32`,
   `port: {ref: u32, node: u32}`, then the 28-byte subscription
-  echo: **48 bytes** (`4 + 4 + 4 + 8 + 28`), not 40. Use
+  echo → 40 bytes. `event ∈ {TIPC_PUBLISHED, TIPC_WITHDRAWN,
-  `'=10I8s'` and assert `struct.calcsize(...) == 48`.
+  TIPC_SUBSCR_TIMEOUT}`.
-  `event ∈ {TIPC_PUBLISHED, TIPC_WITHDRAWN,
-  TIPC_SUBSCR_TIMEOUT}`. Python exposes `TIPC_WAIT_FOREVER` as
-  `-1`, so mask it with `& 0xFFFF_FFFF` before packing an
-  unsigned `I`.
 - **trio shape** — this is where the "nearly-functional,
   modern-async" style pays off; expose it as an `@acm` yielding
   a `trio` receive-channel of typed events, *not* a class:
@@ -576,22 +538,14 @@ async def open_topology_events(
   `kind: Literal['published','withdrawn','timeout']`,
   `addr: TIPCAddress`, `node: int`, `ref: int`. One
   `trio.lowlevel`-free implementation: a nursery-spawned reader
-  task doing `await sock.recv(48)` in a loop. The feed is
+  task doing `await sock.recv(40)` in a loop and
-  authoritative and may neither block the socket reader nor drop
+  `send_nowait()`ing decoded events, with the `@acm` closing the
-  transitions silently. Use `send_nowait()` and, on
+  socket on exit → reader gets `ClosedResourceError` → cancel
-  `trio.WouldBlock`, raise a dedicated
+  scope collapses. Standard `tractor` `@acm` discipline.
-  `TIPCNameEventOverflow` that aborts the subscription and tells
+- **consumer**: `tractor/discovery/_registry.py` gains an
-  the consumer to resubscribe and rebuild its view. A timeout
+  optional "watch" mode so a registrar (or any actor) can keep
-  event is delivered once and then closes the channel. The
+  a live view of the actor set without polling. Sketch the
-  `@acm` cancels its reader before closing the fd so teardown
+  integration in the follow-up issue; do not wire it in PR 1.
-  cannot race a retried `recv()` into `EBADF`.
-- **scope**: topology events carry no publication scope. Use an
-  explicit unknown-scope sentinel and keep the resulting address
-  non-dialable; never copy caller/subscription context into
-  supposedly observed data.
-- **consumer**: [#496] owns the optional watch mode and the
-  decision whether the feed subsumes or merely accelerates
-  existing registrar bookkeeping.
 - **`SOCK_SEQPACKET` is fine here** because this socket never
   goes through `MsgpackTransport` — it's a plain trio socket
   used with `recv()`. The contract's "`SOCK_STREAM` only"
@@ -607,16 +561,14 @@ async def open_topology_events(
 2. `tractor/ipc/_tipc.py`: `TIPCAddress` + `is_tipc_available()`
    predicate + `start_listener()`. No transport yet.
    Tests: address round-trip (`unwrap`/`from_addr`/`wrap_address`),
-   `get_random()` distribution, bind/listen + `SO_ACCEPTCONN`
+   `get_random()` uniqueness, bind/listen + `SO_ACCEPTCONN`
    tolerance, `EAFNOSUPPORT` → actionable `ConnectionError`.
 3. `MsgpackTIPCStream` + `connect_to()` + `get_stream_addrs()`.
    Test: two `trio` tasks in one proc exchange a msg over
    `Msgpack` framing (no `tractor` runtime).
-4. registration tables (contract §2 items 1-8 and 10) +
+4. registration tables (contract §2 items 1-6, 9) +
    `pyproject.toml` mark/extra. Test: full suite under
-   `--tpt-proto tipc` (§7.3). Keep TIPC in the conservative remote
+   `--tpt-proto tipc` (§7.3).
-   preference tier until a follow-up implements and tests contract
-   item 9's node-scope locality policy.
 5. maddr support (`str` form + prefix special-case) + docs.
 6. `open_topology_events()` @acm + its tests (layer B).
 7. docs page + `docs/` example.
@@ -637,8 +589,6 @@ def is_tipc_available() -> bool:
     the `tipc` module is loaded.
 
     '''
-    if sys.platform != 'linux':
-        return False
     try:
         socket.socket(socket.AF_TIPC, socket.SOCK_STREAM).close()
         return True
@@ -646,22 +596,17 @@ def is_tipc_available() -> bool:
         return False
 ```
 
-Do not permanently memoize the result: `modprobe tipc` and module
+Cache it in a module global (it can't change without a
-removal can change it during a long-lived process. Probe once per
+`modprobe`, and a cold call costs a syscall). Pure predicate, no
-runtime startup, or use an explicitly refreshable cache whose
+side effects, no logging.
-owner invalidates it after module-management operations. The
-predicate itself remains side-effect-free and silent.
 
 ### 7.2 gating
 
-- `pytest.mark.tipc` registered in
+- `pytest.mark.tipc` registered in `pyproject.toml`.
-  `_testing/pytest.py::pytest_configure()` via
+- module-level
-  `config.addinivalue_line()`, where this repo declares its other
+  `pytestmark = pytest.mark.skipif(not is_tipc_available(),
-  custom marks. Do not invent a `pyproject.toml` marker table.
+  reason='`tipc` kernel module not loaded (`modprobe tipc`)')`
-- keep pure address, serialization, and topology-codec tests
+  in `tests/ipc/test_tipc.py`.
-  runnable on every host. Apply a shared `requires_tipc` marker
-  only to tests that create sockets or otherwise touch the kernel;
-  do not module-skip `tests/ipc/test_tipc.py`.
 - `--tpt-proto tipc` with no module must fail **loudly and
   early** with the actionable message, not with 400 confusing
   timeouts. Add the check to the `tpt_protos` fixture's existing
@@ -677,29 +622,23 @@ predicate itself remains side-effect-free and silent.
   `sudo modprobe tipc` in a `before` step. GH's
   `ubuntu-latest` runners do allow `modprobe tipc` (the module
   ships with the standard Ubuntu kernel package); verify in a
-  throwaway workflow before wiring the matrix. #493's TIPC leg
+  throwaway workflow before wiring the matrix. If it turns out
-  is now blocking. If runners cease permitting the module load,
+  to be unavailable, fall back to a container job with
-  fix the environment or use a suitable container rather than
+  `--privileged`/`--cap-add NET_ADMIN`, and mark the job
-  silently restoring `continue-on-error`.
+  `continue-on-error` until it's proven stable.
 - cross-node TIPC (bearer) cannot be CI'd; cover it with a
   documented manual smoke test in the docs page, in the style
   of gh #482's LAN examples.
 
 ### 7.4 backend-specific tests worth writing
 
-- **known-name publication/resolution**: bind a listener on
+- **name-publication is discovery**: bind a listener on
   `(stype, inst)`, then from a second task `connect()` by name
   and assert it lands — *without* any `tractor` registrar.
-- **`get_random()` distribution**: 10k `get_random()` calls with
+- **`get_random()` collision resistance**: 10k `get_random()`
-  no live runtime. Do **not** assert 10k distinct values: the
+  calls with no live runtime → 10k distinct `_instance`s.
-  no-runtime seeds and outputs are both only 32 bits. Including
+  (This is the silent-crosstalk risk from §2.3; if the 4-byte
-  duplicate seeds plus distinct-seed hash collisions puts the
+  digest ever collides in this test, escalate to §9.)
-  modeled chance of at least one duplicate near 2.3% for 10k
-  calls. #493 uses `>= n - 2` (modeled probability of more than
-  two collisions around `2e-6`) and separately proves
-  `instance_from_seed()` is a pure function. Also hold actor
-  name/PID fixed while varying only `Aid.uuid` to prove live
-  actors seed from `Aid.uid`.
 - **round-robin surprise**: two listeners bound to the *same*
   `(stype, inst)` both succeed (TIPC allows it) and connects
   distribute. Assert the observed behaviour and reference it
@@ -739,61 +678,14 @@ single best demo this backend has; lead with it.
 
 ## 9. Known risks + escalations
 
-- **Instance collision / silent crosstalk remains unresolved.**
+| risk | mitigation |
-  `Aid.uid` seeding and §7.4 tests reduce and measure risk, but
+| --- | --- |
-  the instance field is still a hard 32 bits. [#501] owns
+| `_instance` hash collision → silent crosstalk (two actors share a service name, TIPC round-robins connects between them) | §7.4 test; if it bites, add a post-bind verification handshake, or bump to a 6-byte digest folded into `(stype_low, instance)` |
-  post-bind verification and recovery. Do not fold bits into
+| kernel/module unavailability everywhere (dev boxes, macOS, CI) | hard gating (§7.2); TIPC is explicitly an *opt-in cluster* transport, never a default |
-  `_stype`: topology can watch only one service type.
+| `getsockname()` returns port-id not name | the `rebind_from_sockname` opt-out (§3.2), landed first |
-- **Concurrent registrar startup can split brain.** Topology can
+| unregistered `/tipc` multiaddr proto | `str` maddr fallback (§4) + upstream track gh #483 |
-  observe duplicate publisher port IDs but cannot elect or fence
+| stale docs (#378 notes tipc.io docs may be out of date) | treat `include/uapi/linux/tipc.h` + `net/tipc/` as the only normative source; cite file+symbol in code comments |
-  a winner; a separate election protocol is required (§5.1).
+| `SOCK_SEQPACKET` topology framing byte-order | probe helper + `?TODO` (§5.2) |
-- **Kernel/module availability is opt-in.** Keep the hard gate in
-  §7.2; TIPC is never the default transport.
-- **A listener sockname is a port ID, not its service name.** Keep
-  the `rebind_from_sockname` opt-out (§3.2).
-- **`/tipc` is not yet a registered multiaddr protocol.** Keep
-  the interim `str` maddr fallback (§4) and upstream gh #483.
-- **The public TIPC docs can be stale.** Treat
-  `include/uapi/linux/tipc.h` and `net/tipc/` as normative and
-  cite file/symbol names in code comments.
-- **A slow topology consumer loses continuity.** Fail fast with
-  `TIPCNameEventOverflow`; resubscribe and rebuild rather than
-  block the reader or retain stale state (§5.2).
-- **TIPC locality preference is not implemented.** Current
-  `_is_local_addr()` handles only UDS and TCP, so node- and
-  cluster-scope TIPC both remain in the conservative remote tier.
-  Add explicit scope-aware policy and multihomed selection tests
-  before claiming node-scope preference (contract §2.9).
-
-### 9.1 remaining constructor/error cleanup
-
-#493 closes the peer-withdrawal race in transport construction,
-but it is not a blanket error-path cleanup. Keep these gaps
-explicit rather than reporting the backend as fully hardened:
-
-- direct `TIPCAddress(...)` construction bypasses
-  `from_addr()` scope normalization; `is_valid` is queried later
-  rather than enforcing validity at construction. Decide whether
-  constructors should reject bad service types/instances/scopes
-  or document direct construction as trusted-internal.
-- `maybe_node`/`maybe_ref` are excluded from `.unwrap()` but, as
-  `msgspec.Struct` fields, still participate in structural
-  equality/hash. If service-name identity must ignore observation
-  metadata, represent or compare it explicitly instead of relying
-  on the current "observed-only" description.
-- `start_listener()` must keep ownership of the raw socket through
-  `bind()`, `listen()` and `SocketListener(...)`. The downstream
-  implementation normalizes bind errors but does not yet wrap the
-  complete listener-construction sequence in close-on-error, so a
-  later setup failure can leak the fd.
-- `_maybe_sockaddr()` currently degrades every `OSError` to an
-  unknown observed address. Narrow that tolerance to expected
-  peer-withdrawal errors (notably `ENOTCONN`) so unrelated bad-fd
-  or programming failures remain visible.
-- error normalization is intentionally required for an
-  unpublished-name `EHOSTUNREACH`, but setup `setsockopt`,
-  listener-constructor, and topology setup failures still need a
-  consistent policy and focused regression tests.
 
 ## 10. Follow-up issue seeds
 
@@ -803,11 +695,9 @@ explicit rather than reporting the backend as fully hardened:
   `py-multiaddr`, then drop our `str`-maddr fallback (§4). Worth
   filing *alongside* the `wg` spec-submission issue so both
   proposals go up together rather than as one-offs.
-- registrar-less discovery fast path via name derivation ([#499],
+- registrar-less discovery fast path via name derivation (§5.1)
-  §5.1)
 - `TIPC_TOP_SRV`-driven push registry in
-  `discovery/_registry.py` ([#496], §5.2)
+  `discovery/_registry.py` (§5.2)
-- post-bind collision verification and recovery ([#501], §9)
 - `TIPC_IMPORTANCE` for the parent<->child lifetime channel
   (§3.3) — genuinely novel supervision QoS, no other backend
   can do it
@@ -815,7 +705,3 @@ explicit rather than reporting the backend as fully hardened:
   for `tractor.trionics` fan-out (explicitly not `MsgTransport`)
 - dual-link resiliency / multi-homing (#378's "hybrid dual link")
   once bearers are scripted in the docs
-
-[#496]: https://github.com/goodboy/tractor/issues/496
-[#499]: https://github.com/goodboy/tractor/issues/499
-[#501]: https://github.com/goodboy/tractor/issues/501

ai/tpt-backends/02_quic_iroh_backend.md

@@ -3,12 +3,6 @@
 Tracks gh [#353]. Prereq reading:
 [`00_shared_backend_contract.md`](./00_shared_backend_contract.md).
 
-**External-fact rule**: every claim here about `iroh`, UniFFI,
-generated bindings, QUIC wire/security behavior, or multiaddr
-support is provisional until the step-0 API-truth pass records a
-source or probe. Tractor/Trio behavior read from this checkout is
-the only locally proven basis for the plan.
-
 **Thesis**: the value of `iroh` over "just QUIC" is
 `NodeId`-addressed, NAT-traversing, relay-fallback endpoints —
 i.e. a `tractor` actor tree that spans hosts *without* a
@@ -55,22 +49,20 @@ relitigate:
   **But**: build it first as the throwaway spike (§6 step 0) to
   de-risk the iroh API surface before writing the bridge.
 
-Version pinning: treat API stability across `iroh` minors as an
+Version pinning: `iroh` moves fast and has had breaking
-**unverified external constraint** until step 0. Pin the version
+API renames across minors. Pin `iroh>=X.Y,<X.Y+1` in a `quic`
-exercised by the spike to `iroh>=X.Y,<X.Y+1` in a `quic` extra,
+extra, and **write down the exact resolved version + the
-and **write down the exact resolved version + generated
+generated `iroh/_uniffi*` module layout** in the module
-`iroh/_uniffi*` module layout** in the module docstring, because
+docstring, because §2 depends on generated-code internals.
-§2 depends on generated-code internals.
 
 **Step 0 of implementation is an API-truth pass**: install the
-pinned `iroh`, inspect both its generated Python and loaded FFI
+pinned `iroh`, `python -c "import iroh; help(iroh)"`, and record
-symbols, and run the throwaway two-process spike. Record in
+in this doc's §1.1 the real names of: endpoint builder, secret
-§1.1 the real names and observed contracts. Every statement
+key type, `connect`/`accept`, bi-stream open/accept, the
-below about `iroh`, UniFFI, Rust callbacks, or generated symbols
+send/recv methods and their exact signatures/return types, and
-is a **step-0 hypothesis**, not a locally proven fact, unless it
+whether they're `async def`. Everything below uses *provisional*
-is copied into the completed API-truth table with a source or
+names and must be reconciled. Do not skip this; do not guess
-probe. Tractor and Trio behavior cited from this checkout is not
+from memory.
-subject to that qualifier.
 
 ### 1.1 API-truth table (fill in during step 0)
 
@@ -87,120 +79,122 @@ subject to that qualifier.
 | send | `await send_stream.write_all(b)` | |
 | recv | `await recv_stream.read(n) -> bytes\|None` | |
 | half-close | `await send_stream.finish()` | |
-| endpoint close + completion | `close()` / `await closed()` | |
-| resolved node address | relay URL + direct socket addrs | |
-| future start/poll callback ABI | generated symbols + args | |
-| future cancel/complete/free | generated symbols + ordering | |
-| callback quiescence guarantee | after poll/complete/free? | |
-| cancellation terminal poll code | generated enum/value | |
-| iroh exception/status taxonomy | per operation | |
 
 ---
 
 ## 2. The `trio`-native uniffi future bridge (`tractor/ipc/_uniffi_trio.py`)
 
-### 2.1 Step-0 generated-ABI gate
+### 2.1 what uniffi actually generates
 
-The expected generated shape is: start an opaque Rust future,
+`uniffi`'s async support does not use asyncio *semantically* —
-poll it with a C callback, cancel through a generated cancel
+it uses asyncio only as the *executor* for a poll loop. The
-symbol, consume its terminal value/status through `complete`,
+generated python for an `async fn` is, in shape:
-then call `free`. The expected callback may arrive on a foreign
-Rust thread. **All of that is external and provisional.** Step 0
-must identify the exact generated driver and prove, from its
-template/source plus probes:
 
-1. the start, poll, cancel, complete, and free signatures for
+1. call `_uniffi_..._<method>(...)` → returns an opaque
-   every return-type family used by `iroh`;
+   `RustFuture` handle (a `void*`/`u64`).
-2. poll result values and whether callbacks can be synchronous,
+2. loop: call
-   concurrent, repeated, or late;
+   `ffi_..._rust_future_poll_<T>(handle, callback, callback_data)`.
-3. which terminal state permits `complete`, when `free` is
+   The callback is a C-ABI fn pointer invoked **from an
-   legal, and when no callback can still reference Python;
+   arbitrary rust thread** with a poll-result code
-4. whether generated callback-data and call-status objects must
+   (`READY`/`MAYBE_READY`).
-   remain alive, and how generated lifting/errors are applied;
+3. the generated glue's callback resolves an
-5. whether one narrow generated async-driver entrypoint can be
+   `asyncio.Future` via `loop.call_soon_threadsafe(...)`; the
-   replaced without importing or requiring an asyncio loop.
+   coroutine awaits it, then re-polls.
+4. on ready: `ffi_..._rust_future_complete_<T>(handle,
+   &call_status)` → the value; then
+   `ffi_..._rust_future_free_<T>(handle)`.
 
-Do not implement from a remembered UniFFI version. If cancel
+**The asyncio dependency is confined to step 3.** That is the
-does not have a documented path to a terminal, safely freeable
+whole insight: the bridge is ~40 lines.
-state, the native Trio bridge fails the spike gate and the first
-backend uses the infected-asyncio fallback.
 
-### 2.2 Cancellation-safe ownership
+### 2.2 the trio version
 
-Do not let the caller task own a raw handle across an `await`.
+```python
-Introduce an actor-scoped `UniffiFutureSupervisor` running in the
+async def await_rust_future(
-dedicated transport nursery specified in §3.2.1. That nursery
+    poll: Callable,      # ffi_..._rust_future_poll_<T>
-must span parent bootstrap, the service nurseries, and final
+    complete: Callable,  # ffi_..._rust_future_complete_<T>
-deregistration. For each call, its operation task owns the
+    free: Callable,      # ffi_..._rust_future_free_<T>
-**entire** generated lifecycle:
+    handle: int,
+    lift: Callable[[Any], Any],
+) -> Any:
+    '''
+    Drive a `uniffi` rust-future to completion on the current
+    `trio` task, bridging rust-thread wakeups via
+    `TrioToken.run_sync_soon()`.
 
-```text
+    '''
-create handle -> poll/callback loop -> complete -> lift/status
+    token = trio.lowlevel.current_trio_token()
-              -> free -> publish result
+    while True:
-                 ^
+        wake = trio.Event()
-                 cancel request uses generated cancel, then follows
+        # NOTE, invoked from a *rust* thread!
-                 the verified terminal poll/complete/free protocol
+        def _cb(_data, poll_code):
+            token.run_sync_soon(wake.set)
+
+        cb = _UNIFFI_FUTURE_CALLBACK(_cb)   # keep a strong ref!
+        poll(handle, cb, 0)
+        await wake.wait()
+        if <poll_code was READY>:
+            break
+    try:
+        status = _UniffiRustCallStatus.default()
+        res = complete(handle, status)
+        _uniffi_check_call_status(status)   # reuse generated helper
+        return lift(res)
+    finally:
+        free(handle)
 ```
 
-The operation task, not the awaiting caller, creates the handle.
+Critical details, each a real bug if missed:
-Creation and insertion in the supervisor's live-operation set
-must have no cancellation checkpoint between them. The operation
-retains strong references to the C callback trampoline, callback
-data, wake state, call status, and handle until step 0 proves all
-callbacks are quiescent and `free` has returned. Use one stable
-callback per operation unless the verified ABI requires a fresh
-one per poll; in either case, retain every potentially callable
-trampoline. Capture `current_trio_token()` in the Trio owner and
-schedule the wake into Trio with `token.run_sync_soon(...)`; the
-foreign callback only stores its poll result and schedules that
-wake.
 
-Caller cancellation is a request, not handle ownership transfer:
+- **`token.run_sync_soon()` is the only trio API callable from a
-
+  foreign thread**, and it is documented as such. Use it; do
-1. the caller sends an idempotent cancel request and waits under
+  *not* use `trio.from_thread.run_sync` (requires a trio thread
-   a short shield for the operation to acknowledge it;
+  context) and do not touch the `Event` directly from the
-2. the owner invokes the generated cancel function exactly once
+  callback.
-   and continues the **verified** poll/complete/free sequence;
+- **the poll code must reach the trio side.** Capture it in a
-3. once caller cancellation is observed, cleanup completion never
+  `nonlocal`/1-slot list written by the callback *before*
-   wins the race by returning a value. After acknowledgement the
+  `run_sync_soon`, since the callback owns the value. Handle
-   caller continues propagating its original Trio cancellation;
+  `MAYBE_READY` by re-polling (the loop above does).
-   if cleanup outlives the grace period it first abandons its
+- **keep the `ctypes` callback object alive** across the await —
-   result channel while the actor supervisor keeps ownership;
+  a GC'd `CFUNCTYPE` trampoline is a segfault. Bind it to a
-4. actor endpoint teardown stops accepting new calls, requests
+  local *and* make sure the local outlives the `poll()` call
-   cancellation of all live operations, and joins the supervisor
+  window.
-   before destroying endpoint/key state.
+- **cancellation.** `await wake.wait()` is a trio checkpoint, so
-
+  a `Cancelled` can fire while rust still owns the future. On
-There is deliberately no `move_on_after(...): free(handle)`
+  cancel we must still `free(handle)` — and per uniffi, the
-path. A timeout proves only that cleanup is slow; it does not
+  correct sequence is to call the generated
-prove that callbacks are quiescent or that `free` is legal. A
+  `ffi_..._rust_future_cancel_<T>(handle)` then continue
-wedged operation therefore remains visible in the supervisor and
+  polling to completion before `free`. Wrap the whole thing so
-can delay graceful actor shutdown; process-level termination is
+  the cancel path does:
-the final escalation, not an unsafe FFI free.
+  `with trio.CancelScope(shield=True): cancel(handle); <drain
-
+  poll loop>; free(handle)`. **Bounded** shield (add a
-Structured-concurrency race to test: caller cancellation may land
+  `trio.move_on_after()` with a module-level constant) so a
-after handle creation, after each poll, during callback delivery,
+  wedged rust future can't make an actor un-cancellable —
-after terminal readiness, during `complete`, and before result
+  `tractor` is SC-first and an unbounded shield here would
-publication. At every checkpoint exactly one operation task owns
+  violate that.
-the handle, exactly one `free` is possible, and the supervisor
+- **`trio.lowlevel.current_trio_token()`** must be captured on
-cannot exit while that task or a callable trampoline remains.
+  the trio side (not in the callback).
 
 ### 2.3 how to apply it to the generated bindings
 
-Do **not** fork/vendor the generated `iroh` Python. Subject to the
+Do **not** fork/vendor the generated `iroh` python. Instead ship
-step-0 gate, ship a *narrow* re-dispatch shim:
+a *narrow* re-dispatch shim:
 
-- write `tractor/ipc/_uniffi_trio.py` with the supervisor and a
+- write `tractor/ipc/_uniffi_trio.py` with `await_rust_future()`
-  `@cm patch_uniffi_for_trio()` that patches only the generated
+  plus a `@cm patch_uniffi_for_trio()` that monkey-patches the
-  async-driver entrypoint recorded in §1.1;
+  generated module's single async-driver entrypoint (in current
+  uniffi that's `_uniffi_rust_call_async` / `_rust_call_async`,
+  one function) to the trio implementation.
 - verify at import time that the expected symbol exists and
   raise a clear, actionable error naming the pinned `iroh`
   version if not. A silent fallback to asyncio would be a
   nightmare to debug.
-- treat every `iroh`/UniFFI upgrade as requiring the step-0 ABI
+- **plan for this to break on `iroh`/`uniffi` upgrades.** Mitigate
-  gate again. Keep a test that drives one trivial call under bare
+  with (a) a unit test that drives one trivial `iroh` async call
-  `trio.run()`, asserts no asyncio loop, and injects cancellation
+  under bare `trio.run()` and asserts no event loop was ever
-  at every lifecycle checkpoint. Point the module docstring at
+  created (`asyncio.get_event_loop_policy()` untouched /
-  the exact generated template/revision mirrored by the shim.
+  `asyncio._get_running_loop() is None`), and (b) a docstring
+  pointing at the uniffi codegen template this mirrors.
 
 If step 0 reveals the generated code is *structurally* hostile
 to this (e.g. `asyncio` imported and used at module scope for
@@ -234,52 +228,18 @@ iroh bi-stream     ==  one `Channel`/`MsgTransport`  -> 1:1
 - `layer_key: int = 4` still (QUIC is L4-ish); note in a comment
   that this backend is really 4+security+multiplex.
 
-**Connection pooling** is actor-endpoint state, never module
+**Connection pooling** is the one place we add state the other
-state. Its key is exactly
+backends don't have: dialing the same peer twice should reuse
-`(local_endpoint_identity, remote_node_id, alpn)`, where local
+the `Connection` and open a second bi-stream. Implement as a
-endpoint identity is the local NodeId derived from the actor key.
+module-level `dict[NodeId, Connection]` guarded by a
-Remote NodeId alone would incorrectly share connections across
+`trio.Lock`... **no** — that's a per-process cache with
-local keys or protocol epochs. Build it over the codebase's
+lifetime/teardown hazards. Instead reuse the codebase's existing
-`maybe_open_context()` idiom only after a concurrency review of
+idiom: `tractor.trionics.maybe_open_context()` keyed on the
-its actual last-user teardown behavior in the implementation
+node-id, which already solves exactly this (one-cached-resource-
-revision. Do not assume an issue reference proves the required
+per-key, refcounted, teardown-on-last-exit) and whose teardown
-ordering.
+semantics were just hardened (gh #488). Use it; do not hand-roll
-
+a cache. Anything concurrency-subtle here should get the
-`acquire_connection()` returns a `ConnectionLease`, not a bare
+`conc-anal` skill run over it.
-connection. An outgoing `QuicMsgStream` owns that entered lease
-for its whole lifetime; `connect_to()` must not exit the cached
-context immediately after `open_bi()`. Exact transfer paths:
-
-- dial/acquire or `open_bi()` failure releases the lease in a
-  shielded `finally` before raising;
-- successful stream construction atomically transfers the lease
-  to `QuicMsgStream` before the first cancellation checkpoint;
-- `send_eof()` closes only the send half and does not release;
-- clean receive EOF closes only the receive half and does not
-  release while the send half remains usable;
-- one guarded terminal-state transition releases exactly once
-  when both halves have become terminal, in either order;
-- `aclose()`, reset, or terminal connection failure closes both
-  halves as applicable and idempotently releases exactly once;
-- a stream queued by `QuicListener` already owns its lease; if
-  never accepted, listener draining closes it and releases it.
-
-After `accept()` returns, the server dispatch path owns the stream
-until a handler task starts and must close it if task start fails.
-The handler then takes ownership, with an outer `finally` that
-calls `stream.aclose()` on normal return, handshake failure, and
-cancellation. Lease release itself is an idempotent pool state
-transition; if last-user connection teardown awaits FFI, the actor
-endpoint's pool supervisor owns that await so cancellation of the
-handler cannot strand the lease.
-
-For inbound connections, the connection-feeder owns a base lease
-while accepting streams and each queued/returned stream gets a
-child lease. The base lease is released only after the accept
-loop ends; the pool closes the connection after the base and all
-stream leases are gone. Reject or deterministically reconcile a
-simultaneous inbound/outbound duplicate for the same full key;
-record the chosen iroh-compatible rule during step 0.
 
 ### 3.2 `IrohAddress`
 
@@ -288,29 +248,29 @@ class IrohAddress(
     msgspec.Struct,
     frozen=True,
 ):
-    _node_id: str
+    _node_id: str                 # 32B ed25519 pubkey, hex or z32
-    _alpn: str
+    _alpn: str = 'tractor/0'      # the bindspace!
-    _relay_url: str|None
+    # optional dial hints; NOT part of identity
-    _direct_addrs: tuple[str, ...]
+    maybe_relay_url: str|None = None
+    maybe_direct_addrs: tuple[str, ...] = ()
 
-    proto_key: ClassVar[str] = 'quic'
+    proto_key: ClassVar[str] = 'iroh'   # ?or 'quic'; see §3.2.1
-    unwrapped_type: ClassVar[type] = tuple
+    unwrapped_type: ClassVar[type] = tuple[str, str]
     def_bindspace: ClassVar[str] = 'tractor/0'
 ```
 
-- **`.unwrap()` is the complete, tagged wire descriptor**:
+- **`.unwrap() -> (node_id_str, alpn_str)`** — a `(str, str)`
-  `('quic', node_id, alpn, relay_url, direct_addrs)`. All values
+  tuple, which is *unambiguously distinct* from
-  are msgpack-native and `direct_addrs` is canonicalized to a
+  `TCPAddress`'s `(str, int)`. But careful:
-  tuple. `from_addr()` requires that exact tag and shape; never
+  `wrap_address()`'s UDS case is
-  infer QUIC from a `(str, str)` pair. This depends on the shared
+  `case (_, filename) if type(filename) is str` — which
-  contract's tagged-address migration and removes the UDS
+  **already catches `(str, str)`**. So the iroh `case` MUST be
-  collision rather than ordering around it.
+  ordered *before* the UDS case and guarded, e.g.
-- The descriptor always carries NodeId, ALPN, and both route-hint
+  `case (str() as nid, str() as alpn) if _is_node_id(nid):`
-  fields. For this discovery-free first backend, `.is_valid`
+  with `_is_node_id()` a cheap length+alphabet check. Add a
-  requires a parseable NodeId, non-empty ALPN, and at least one
+  regression test asserting a UDS `(dir, filename)` pair still
-  relay URL or direct address. Whether NodeId-only dialing works
+  wraps to `UDSAddress` — this is the exact "wrong transport
-  through optional iroh discovery is a step-0 API check and is
+  loaded" hazard `_addr.py:214` warns about.
-  not part of the first implementation.
 - `.bindspace` → `self._alpn`. This is the honest analogue:
   the ALPN is the set of endpoints willing to talk to you, and
   two `tractor` deployments sharing an iroh network are
@@ -318,80 +278,39 @@ class IrohAddress(
   separated by directory. Include a `tractor` version/proto
   epoch in the default ALPN so incompatible runtimes can't
   handshake.
+- `.is_valid` → node-id parses, alpn non-empty.
+- **`get_root()` is the hard one.** There is no
+  well-known-port analogue: an iroh node id is a *keypair*, so
+  "the host's default registrar addr" requires a *persisted
+  secret key*. Design:
+  - the root/registrar's secret key lives at
+    `get_rt_dir() / 'iroh_registrar.key'` (0600), created on
+    first use.
+  - `get_root()` must stay **pure and import-time-safe**
+    (contract §2.3: `_default_lo_addrs` is built at import!).
+    So `get_root()` *reads* the key file if present and
+    otherwise returns an `IrohAddress` with
+    `_node_id=''`/sentinel, and the **generation** happens in
+    an explicit sibling — `ensure_registrar_key() ->
+    IrohAddress` — called from the listen path. Pure getter,
+    explicit setter; do not smuggle key generation into
+    `get_root()`.
+  - this almost certainly means `_default_lo_addrs` must become
+    lazy for this backend. **Land that refactor as its own prep
+    commit** (a `default_lo_addrs()` that computes per-call
+    instead of the import-time dict) — it also unblocks plan
+    03's netns-scoped defaults.
+- `get_random()`: generate a fresh `SecretKey` per subactor and
+  return its node-id. Note this runs post-fork pre-listen
+  (contract §4) and costs an ed25519 keygen (~µs, fine). The
+  *secret* can't live in a frozen `Address`, so it must be
+  stashed where the listen path can find it: a module-level
+  `dict[node_id, SecretKey]` populated by `get_random()` and
+  consumed+popped by `start_listener()`. Ugly but honest;
+  document it and note the alternative (thread the key through
+  `Endpoint`) as a follow-up.
 
-### 3.2.1 One actor endpoint and key
+#### 3.2.1 `proto_key`: `'iroh'` vs `'quic'`
-
-Add an actor-scoped `QuicActorEndpoint` resource containing the
-secret key, one bound iroh endpoint, the UniFFI supervisor, the
-connection pool, and its latest resolved `IrohAddress`. It cannot
-live in `_service_tn`: a child dials its parent before that nursery
-opens, while final deregistration may dial after it closes.
-
-Add a dedicated `transport_tn` around the complete actor runtime:
-the task that opens this nursery must start the complete
-`async_main` sequence as a **child** of it and wait for that child.
-That makes `transport_tn` an ancestor of every parent-dial,
-service, and deregistration caller, satisfying
-`maybe_open_context(tn=transport_tn)` rather than asking the
-nursery-opening task to use its own child nursery. The child keeps
-the nursery around `_root_tn` and `_service_tn`, performs final
-deregistration while it remains open, then returns so the owner can
-close the transport resource and nursery. Root startup needs the
-equivalent outer owner around actor construction, service, and
-teardown. If this shape cannot be preserved, the connection pool
-must stop depending on `maybe_open_context()`'s ancestor-nursery
-contract. No path creates a second endpoint for the actor.
-
-The child currently receives transport configuration only in the
-`SpawnSpec` sent over its already-open parent channel. QUIC cannot
-derive its local key, ALPN, or requested bind policy from that late
-message. Add a small msgpack/pickle-native
-`ChildTransportBootstrap` to every process-launch path. It carries
-the selected protocol and the QUIC-local key reference/generation
-policy, ALPN, relay policy, and requested bind constraints. It is
-available before `_from_parent()`; the later `SpawnSpec` repeats
-the public configuration and startup rejects any mismatch. Root
-actors derive the same bootstrap record directly from
-`open_root_actor()` inputs before address selection.
-
-With that prep in place, the order is:
-
-1. consume the launch-time bootstrap record, select one key
-   (persisted and explicitly provisioned for a registrar, fresh
-   for an ordinary actor), and construct/bind the endpoint in the
-   transport owner task;
-2. await the step-0-verified address-ready API and build a valid
-   descriptor from the endpoint's NodeId, ALPN, relay URL, and
-   direct addresses;
-3. only then dial `_from_parent()` through this endpoint;
-4. start `QuicListener` over this endpoint's accept API;
-5. publish the resolved descriptor as `Endpoint.addr` and
-   `Actor.accept_addrs` before parent/registrar registration;
-6. after service nurseries close, keep the endpoint available for
-   deregistration; then close listeners and streams, drain
-   connection leases and FFI operations, close/join the endpoint,
-   and release key state.
-
-`IrohAddress.get_random()` is therefore a descriptor lookup on
-the active actor transport resource, not key generation. Broaden
-the shared `get_random()` contract for resource-backed transports
-and make root/subactor address selection consume the bootstrap
-resource instead of calling it before that resource exists. Do not
-hide a secret in a module-level side table. Calls without an active
-resource fail clearly rather than allocating an unowned key.
-
-`get_root()` never returns an empty/sentinel NodeId. Make default
-addresses lazy, and have QUIC load a provisioned public registrar
-descriptor. Registrar provisioning writes its secret separately
-with mode 0600 and writes the matching complete public descriptor
-atomically; endpoint startup verifies the derived NodeId. If no
-descriptor exists, default QUIC registrar discovery fails with an
-actionable configuration error. Automatic first-process election
-is deferred until a safe key-file locking and endpoint-binding
-protocol is proven; key generation never occurs in the listen
-path.
-
-#### 3.2.2 `proto_key`: `'iroh'` vs `'quic'`
 
 Use **`'quic'`** for the `proto_key`/`--tpt-proto` name and
 name the module `_quic.py`, with `iroh` as the *implementation*.
@@ -438,7 +357,7 @@ class QuicMsgStream(trio.abc.HalfCloseableStream):
     '''
     tpt_key: ClassVar[MsgTransportKey] = ('msgpack', 'quic')
 
-    def __init__(self, conn, send, recv, lease) -> None: ...
+    def __init__(self, conn, send, recv) -> None: ...
     async def send_all(self, data: bytes) -> None: ...
     async def wait_send_all_might_not_block(self) -> None: ...
     async def receive_some(self, max_bytes: int|None = None) -> bytes: ...
@@ -460,38 +379,17 @@ already exists in `_transport.py` and must keep working):
   absent, so the `raise_on_report` branch at
   `_transport.py:290` stays quiet).
 - `send_all()` on a closed peer → `trio.BrokenResourceError`.
-- honour Trio's one-task-per-direction rule with public,
+- honour `trio`'s one-task-per-direction rule: guard with
-  implementation-local guards that raise
+  `trio._util.ConflictDetector` equivalents (or just document +
-  `trio.BusyResourceError`; do not depend on `trio._util`.
+  assert), because `MsgpackTransport` already serializes sends
-  `MsgpackTransport` already serializes sends, while receives are
+  with a `StrictFIFOLock` but recvs are single-task by
-  single-task by construction.
+  construction.
 - **buffering**: if iroh's `read()` doesn't support
   "read up to n", `receive_some()` must maintain an internal
   leftover buffer. Note `MsgpackTransport` wraps us in
   `tricycle.BufferedReceiveStream` anyway, so `receive_some()`
   just needs *some* nonzero-progress contract.
 
-Centralize exception translation at every iroh/UniFFI boundary;
-no generated exception may escape into `Channel` or server code.
-Step 0 must record actual exception classes/status payloads and
-build an exhaustive operation-specific mapping:
-
-| observed condition | adapter result |
-| --- | --- |
-| receive clean EOF | `b''` |
-| local stream/listener/endpoint already closed | `trio.ClosedResourceError` |
-| concurrent same-direction operation | `trio.BusyResourceError` |
-| peer reset, stopped stream, lost connection | `trio.BrokenResourceError` |
-| dial rejected or no usable route | `ConnectionRefusedError` or `ConnectionError` |
-| caller's Trio deadline/cancellation | preserve Trio cancellation semantics |
-| unexpected FFI status/panic | chained `RuntimeError` identifying operation and pinned version |
-
-Preserve the original exception as `__cause__`, but sanitize
-messages so `_transport.py` sees stable Trio/Tractor categories,
-not version-specific iroh text. Endpoint accept failure becomes a
-listener `BrokenResourceError`; normal endpoint shutdown becomes
-`ClosedResourceError`. Add one test per observed step-0 status.
-
 ```python
 class QuicListener(trio.abc.Listener):
     '''
@@ -504,63 +402,41 @@ class QuicListener(trio.abc.Listener):
     async def aclose(self) -> None: ...
 ```
 
-The accept-side subtlety is fan-out: one actor transport accepts
+The accept-side subtlety: `trio.abc.Listener.accept()` yields
-connections and each connection accepts streams, while
+one stream per call, but iroh gives us *connections* which then
-`Listener.accept()` returns one stream. Give **each** listener a
+yield *streams*. So `QuicListener` needs an internal
-supervisor task started with
+`trio.MemoryReceiveChannel[QuicMsgStream]` fed by a background
-`await server_ep.listen_tn.start(...)`.
+task-pair (one task accepting connections, one per connection
-That task creates and owns a cancel scope, a child nursery for the
+accepting bi-streams). `trio.abc.Listener` has no nursery, so:
-endpoint feeder plus per-connection feeders, a guarded stream
+make the listener **constructed by an `@acm`** that owns the
-queue, and a completion event.
+nursery, and have `start_listener()` be that `@acm`'s driver.
-`start_listener(addr=, server_ep=, actor_tpt=)` does not return
-until the supervisor has reported all of those ready.
-Do not borrow an implicit parent nursery or spawn feeders lazily
-from `accept()`.
 
-The queue is a guarded `deque`, not an unowned memory-channel
+⚠️ this collides with `Endpoint.start_listener()` being a plain
-buffer. A feeder transfers a fully constructed, lease-owning
+`async def` returning a listener. Two options:
-stream into it only while the listener is open; if close wins the
+- **(a)** hang the nursery off the `Endpoint`'s existing
-race, the feeder closes the stream itself. `accept()` atomically
+  `listen_tn` — `_serve_ipc_eps()` already creates `listen_tn`
-pops one item or waits on the queue condition. Once close is
+  and passes it into every `Endpoint` (`_server.py:1063-1074`),
-marked and the queue is empty, it raises
+  and `Endpoint.listen_tn` is right there. So
-`trio.ClosedResourceError`.
+  `start_listener()` can `self.listen_tn.start_soon(...)` the
+  acceptor tasks. **Recommended**: no upstream signature change,
+  correct lifetime (dies with the ep group), and it's why
+  `listen_tn` is on the struct in the first place.
+- (b) change `start_listener()` to a `@acm`. Bigger blast
+  radius; only if (a) proves insufficient.
 
-`QuicListener.aclose()` is idempotent and has this exact order:
+Since `start_listener()` is called via
-
+`inspect.getmodule(addr)` with only `addr=` (contract §1.3),
-1. under the queue guard, mark closed and wake all `accept()`
+option (a) needs the `Endpoint` itself. Either add `ep=` to the
-   waiters without a checkpoint between the state change and
+module-level `start_listener()` call signature (all backends
-   notification;
+ignore it except quic → small upstream change, do it as part of
-2. cancel the listener-owned supervisor scope;
+the prep PR and make it keyword-only with a default) or have
-3. the supervisor's shielded `finally` joins the endpoint and all
+`QuicListener.accept()` lazily spawn via
-   connection feeders, atomically detaches the queue, closes every
+`trio.lowlevel.current_task().parent_nursery` (**rejected** —
-   queued stream, releases their leases, and closes the queue;
+fragile, implicit). Do the explicit `ep=` kwarg.
-4. only after that finalizer finishes, the supervisor sets its
-   completion event;
-5. `aclose()` waits under a shield for that event and returns;
-   concurrent closers wait for the same event.
-
-The same supervisor finalizer runs if its parent nursery is
-cancelled before someone calls `aclose()`. This makes the
-supervisor, not an arbitrarily cancelled caller, the sole final
-cleanup owner. Test cancellation at feeder accept, stream
-construction, queue transfer, `accept()` wakeup, and each close
-checkpoint; no feeder may outlive the listener and no queued
-lease may survive completion.
-
-This needs two explicit, typed references in the module-level
-listener call: `server_ep=` is the IPC server `Endpoint` that owns
-`listen_tn`, while `actor_tpt=` is the already-open
-`QuicActorEndpoint` whose iroh accept API supplies connections.
-Store `actor_tpt` on the server endpoint during actor transport
-bootstrap and pass both keyword-only arguments; socket backends
-ignore `actor_tpt`. `Endpoint.start_listener()` then stores the
-listener's already-resolved address instead of calling
-`getsockname()`.
 
 ### 3.4 `maddr`
 
-Expected multiaddr spellings for direct QUIC and relay routes are
+Multiaddr already standardizes the pieces:
-**step-0 verification items**, not assumptions:
 
 ```
 /ip4/<h>/udp/<p>/quic-v1                     # direct
@@ -568,16 +444,19 @@ Expected multiaddr spellings for direct QUIC and relay routes are
 /dns/<relay-host>/tcp/443/tls/ws/p2p/<node>  # relay-ish
 ```
 
-- Do not emit NodeId alone in the first backend: without enabled
+- primary form: `/p2p/<node-id>` alone is a legal maddr and is
-  discovery it would discard the route required by the complete
+  the *only* required component for iroh dialling — relay +
-  `IrohAddress`. `mk_maddr()` must preserve NodeId, ALPN, relay
+  direct addrs are discovery hints. So `mk_maddr()` emits
-  URL, and all direct addresses, or return a canonical Tractor
+  `/p2p/<node_id>` and, when known, prefixes the direct
-  string form that does until a multiaddr grammar can round-trip
+  `/ip4/../udp/../quic-v1/`.
-  every field.
+- `/p2p/` values are multihash-encoded peer ids; an iroh node-id
-- Verify whether an iroh NodeId can losslessly map to `/p2p/`.
+  is a raw ed25519 key. Converting requires the identity
-  If not, use a tractor-local `/iroh/<node-id>` segment rather
+  multihash + libp2p key protobuf wrapper. **Decide**: emit the
-  than pretending to be a libp2p peer-id. This needs upstream
+  raw node-id under a *tractor-local* `/iroh/<node-id>` segment
-  registration, on the same track as `wg`/`tipc` (gh #483).
+  (needs upstream registration, same track as `wg`/`tipc`,
+  gh #483) rather than pretending to be a libp2p peer-id we
+  can't round-trip. Return the `str` form until upstream lands
+  (`MsgTransport.maddr` is `Multiaddr|str`).
 - this backend is the strongest argument for gh #443's
   **tunnelled/composed maddr** item: `/ip4/../udp/../quic-v1/..`
   *is* a composed stack. Cross-reference plan 03 §5 so the two
@@ -587,25 +466,24 @@ Expected multiaddr spellings for direct QUIC and relay routes are
 
 ## 4. Discovery integration
 
-- The registrar stores the complete `IrohAddress`, not only a
+- iroh's node-id addressing means the `tractor` registrar can
-  NodeId. Registration is forbidden until endpoint address
+  hold `IrohAddress`es that are **reachable from anywhere** with
-  resolution has produced that descriptor. If route hints change
+  no port-forwarding — that is the headline feature. The
-  later, dynamic re-registration is a follow-up; the spike uses
+  registrar itself works unchanged.
-  the pre-registration snapshot.
+- iroh has its own discovery (DNS/pkarr/mdns). **Out of scope**;
-- Optional iroh discovery mechanisms and their names/capabilities
+  note in the follow-up that `tractor.discovery` could
-  are step-0 verification items and out of scope for the first
+  eventually delegate to it, which would be the direct analogue
-  backend. No NodeId-only reachability claim is made.
+  of plan 01's TIPC-topology idea.
-- Relay configuration belongs to `QuicActorEndpoint` creation,
+- relay servers: default to n0's public relays for the demo,
-  not `start_listener()`, because dialing and listening reuse the
+  document self-hosting (docs.iroh.computer's dedicated-infra
-  same endpoint. The demo's relay choice and self-hosted option
+  page is linked from #353), and make the relay set a
-  are selected only after step 0 verifies the pinned API.
+  `start_listener()` kwarg.
 
 ## 5. Security note
 
-The transport-security and NodeId-authentication properties of the
+QUIC is TLS-1.3-always and iroh authenticates by node-id, so
-pinned iroh stack are **step-0 documentation-verification items**.
+this backend is the first `tractor` transport with real
-Claim only the properties supported by that version's source and
+transport security and peer authentication. Two things follow:
-docs. Two design consequences remain:
 1. an **allowlist hook** — an actor should be able to reject
    inbound connections from unknown node-ids *before* the
    `Aid` handshake. Natural home: a predicate kwarg on
@@ -619,72 +497,59 @@ docs. Two design consequences remain:
 
 0. **spike (throwaway, not committed)**: drive iroh under
    `trio-asyncio`/`tractor.to_asyncio`, echo bytes over a
-   bi-stream between two processes. Fill §1.1 with generated ABI,
+   bi-stream between two procs. Fills in §1.1. Timebox it.
-   endpoint resolution, close/join, and error observations. Probe
+1. prep PR: annotation widening + `rebind_from_sockname` gate +
-   cancel at every generated lifecycle phase. Timebox it and use
+   `transport_from_stream()` `tpt_key` dispatch + `ep=` kwarg on
-   the fallback if any mandatory ownership fact stays unknown.
+   `start_listener()` + lazy `default_lo_addrs()`. **No new
-1. prep PR: tagged address migration, annotation widening,
+   backend.** Full suite green on tcp *and* uds.
-   non-socket listener reconciliation, `tpt_key` dispatch,
+2. `_uniffi_trio.py` + its tests (drive one iroh async call
-   typed `server_ep=`/`actor_tpt=` listener inputs, and lazy
+   under bare `trio.run()`; assert no asyncio loop; assert
-   default addresses. **No new backend.** Keep tcp and uds
+   cancellation frees the future).
-   behavior unchanged.
+3. `QuicMsgStream` + tests against a *loopback* iroh endpoint
-2. bootstrap prep: pass `ChildTransportBootstrap` through every
+   pair in one process (no `tractor` runtime): send/recv, clean
-   process-launch path and add the transport nursery around child
+   EOF → `b''`, reset → `BrokenResourceError`, use-after-close
-   parent-dial, service, deregistration, and teardown. Resolve the
+   → `ClosedResourceError`.
-   endpoint address before registration. Add no iroh-specific
+4. `QuicListener` + `start_listener()` + `IrohAddress` +
-   global state.
+   key-file mgmt.
-3. `_uniffi_trio.py` supervisor + lifecycle fault-injection tests:
+5. `MsgpackQuicStream(MsgpackTransport)` + `connect_to()` +
-   no asyncio loop, one owner/complete/free, callback retention,
+   `maybe_open_context()` connection pooling.
-   bounded caller handoff, and joined durable cleanup.
+6. registration tables + `--tpt-proto quic` + full suite.
-4. `QuicActorEndpoint` + provisioned registrar descriptor +
+7. maddr + docs + a two-host example (pairs with #482's format).
-   loopback direct-address tests; prove one endpoint handles dial,
-   listen, address lookup, and ordered teardown.
-5. `QuicMsgStream` + exhaustive error-normalization and lease
-   release tests against the loopback endpoint pair.
-6. `QuicListener` supervisor + cancellation-at-every-checkpoint
-   tests, including queued-stream draining and feeder joins.
-7. `MsgpackQuicStream`, full-key connection pooling, registration
-   tables, and `--tpt-proto quic`; then run the full suite.
-8. routable maddr/string form + docs + a two-host example (pairs
-   with #482's format).
 
 ## 7. Testing
 
 - capability predicate `is_quic_available()` → `iroh` importable
-  *and* every step-0-recorded driver symbol present at the pinned
+  *and* the uniffi driver symbol present at the pinned version.
-  version. Same `pytest.fail`-early hook as plan 01 §7.2.
+  Same `pytest.fail`-early hook as plan 01 §7.2.
 - **the acceptance bar is the same**: whole suite green under
   `--tpt-proto quic`. Expect this to shake out real bugs in the
   adapters (esp. teardown ordering and `TransportClosed`
   classification) — that's the point.
-- Measure endpoint bind and first-connect latency in step 0; do
+- expect to need **timeout headroom**: iroh endpoint bind +
-  not assume a multiplier. Before changing a deadline, rule out
+  first connect (relay discovery) is orders of magnitude slower
-  the project's CPU-throttle false-positive, then prefer one
+  than a UDS bind. Before touching any test deadline, rule out
-  per-proto harness multiplier over individual-test edits.
+  the CPU-throttle false-positive (see the project's
-- Use the step-0-verified relay-disable configuration with direct
+  `env_cpu_throttle_masquerades_as_regression` note); then, if
-  loopback addresses for default CI. Mark separately verified
+  real, add a per-proto timeout multiplier to the test harness
-  relay tests `pytest.mark.net` and keep them out of default CI.
+  rather than editing individual tests.
-- leak checks: assert the actor has one key/endpoint, every FFI
+- a no-network test mode: iroh with relays disabled +
-  operation completed/freed once, all listener feeders joined,
+  loopback direct addrs only, so CI doesn't depend on n0's
-  all queued streams closed, every connection lease released,
+  infra. **Make this the default in CI**; mark the relay tests
-  and endpoint close completion observed before actor teardown.
+  `pytest.mark.net` and keep them out of the default run.
-- address-ordering check: block registration until a descriptor
+- leak checks: assert every `SecretKey`/`Endpoint` is closed on
-  with NodeId, ALPN, and at least one route is published; reject
+  actor teardown (an `Endpoint` left open holds UDP sockets and
-  sentinel, NodeId-only, and post-registration mutation cases.
+  relay connections; a leak here shows up as hung tests, not
+  errors).
 
 ## 8. Risks
 
 | risk | mitigation |
 | --- | --- |
 | uniffi codegen internals shift on upgrade | pinned minor, symbol assertion at import, the "no asyncio loop" test, documented fallback to `to_asyncio` |
-| callback wakeup/lifetime semantics differ from the hypothesis | step-0 source + probe gate; retain callback/data through verified quiescence; durable owner; never timeout-free |
+| rust-thread callback → trio wakeup mishandled (segfault / lost wakeup / un-cancellable task) | strong ref on the ctypes trampoline; `run_sync_soon` only; **bounded** shielded cancel-drain; run the `conc-anal` skill over the bridge |
-| cancelled foreign future never reaches a freeable state | bounded caller handoff to visible actor supervisor; joined graceful shutdown or process-level escalation; never speculative free |
 | `iroh` wheel availability for 3.13/3.14 on linux+macos | verify in step 0; if missing, that alone may force the `aioquic` fallback |
-| endpoint or route resolution is not ready before parent dial/registration | actor endpoint bootstrap barrier; publish only a complete resolved descriptor |
+| QUIC latency/jitter destabilizes the existing suite's timing assumptions | per-proto timeout multiplier, relay-less CI mode |
-| connection closes while a stream still uses it | full-key pool + stream-held leases + exact-once release tests |
+| `(str, str)` unwrapped form collides with UDS in `wrap_address()` | guarded case ordered first + explicit regression test (§3.2) |
-| listener close strands feeder tasks or queued streams | listener-owned scope/completion event; cancel, join, drain, then return |
-| QUIC latency/jitter destabilizes suite timing assumptions | measure first; per-proto multiplier only if demonstrated; relay-less CI mode |
-| address tuple collides with another backend | required `'quic'` tag and exact-shape dispatch |
 | scope creep into iroh's docs/blobs/gossip crates | this backend is `Endpoint`+`Connection`+bi-streams only; anything else is a separate issue |
 
 ## 9. Follow-up issue seeds

ai/tpt-backends/03_wg_tunnel_bindspace.md

@@ -33,7 +33,7 @@ onto `trio` as the library's sans-io layer allows.
   by multiformats/py-multiaddr#107 and gh #483.
 - so **today's deployable story is declarative**: run `wg-quick`
   out-of-band, parse the maddr, strip to the overlay
-  `(host, port)`, verify the pubkey in its host-specific role,
+  `(host, port)`, verify the pubkey against the live tunnel,
   hand the overlay addr to `registry_addrs=`/`tpt_bind_addrs=`.
   #482 already contains working example code for exactly this.
 - `Address.namespace` exists in the Protocol
@@ -82,10 +82,9 @@ does not create a new address type.** Two candidate encodings;
   the tunnel spec. The wrapper is stripped (`→ .overlay`) at the
   moment of bind/connect.
   - ⚠️ `is_wrapped_addr()` (`_addr.py:194`) tests
-    `type(addr) in _address_types.values()` — the build-registered
+    `type(addr) in _address_types.values()` — a `bidict` of
-    protocol-key-to-address-type registry. `TunnelledAddress`
+    proto_key→type. `TunnelledAddress` isn't in it and must not
-    isn't in it and must not be (it has no transport of its own).
+    be (it's not 1:1 with a proto). So either add an explicit
-    So either add an explicit
     `isinstance(addr, TunnelledAddress)` clause there, or give
     the wrapper a marker and test structurally. Do the former;
     it's two lines and honest.
@@ -104,7 +103,7 @@ class WGTunnelSpec(
     msgspec.Struct,
     frozen=True,
 ):
-    pubkey: str               # std-base64 `wg(8)` form
+    peer_pubkey: str          # std-base64 `wg(8)` form
     iface: str = 'wg0'
     netns: str|None = None
     # layer-C-only fields, unset in layer A
@@ -192,10 +191,9 @@ Observed protocol-name lists, for writing the `match`:
   ...))` w/ the bearer recorded in the spec.
 - keep the existing 2-proto cases byte-identical; add the new
   case *after* them.
-- layer A rejects more than one `/wg/` segment. Its wrapper stores
+- nesting (wg-in-wg) falls out of `.decapsulate_code()` cutting
-  one bearer, key, and overlay, so accepting wg-in-wg would
+  at the *last* occurrence — peel repeatedly rather than
-  silently misrepresent the maddr. Nested tunnel support needs a
+  recursing through a bespoke splitter.
-  different data shape and belongs in a later layer.
 - `mk_maddr()` inverse for `TunnelledAddress` is just
   `.encapsulate()` composition; don't rebuild `str`s by hand.
 - **pending an upstream release**: py-multiaddr#108 is merged, so
@@ -217,21 +215,12 @@ Port #482 §2's helpers into `tractor/discovery/_tunnel.py` as
 ```python
 def parse_wg_maddr(maddr: str) -> TunnelledAddress: ...   # pure
 def wg8_pubkey(multibase_key: str) -> str: ...            # pure
-async def verify_wg_key(
+def verify_wg_peer(spec: WGTunnelSpec) -> bool: ...       # impure probe
-    spec: WGTunnelSpec,
-    role: Literal['local', 'peer'],
-    inspection: str|None = None,
-) -> bool: ...  # impure probe
 ```
 
-In layer A `verify_wg_key()` may shell out to role-specific
+In layer A `verify_wg_peer()` may shell out (`wg show <if>
-`wg show <if> public-key|peers` queries, but it must be a *single*
+peers`), but it must be a *single* function so layer B swaps
-async, time-bounded function so it never blocks trio's run thread
+only its body. Never call it implicitly from
-and layer B swaps only its body. It verifies key presence only,
-not `Endpoint`, `AllowedIPs`, handshake state, or routing. Never
-run `tractor` as root: privileged inspection stays a separate
-step whose public-key output can be passed as `inspection`. Never
-call it implicitly from
 `wrap_address()`/`parse_maddr()` — parsing must stay pure and
 side-effect-free; verification is the *caller's* explicit step
 (and later, the bindspace `@acm`'s).
@@ -292,7 +281,7 @@ Three integration options, in increasing trio-nativeness:
 - (3) reimplement the codecs. Never.
 
 **Recommended split**: ship (1) first so layer B is a small,
-reviewable, behaviour-preserving swap of `verify_wg_key()`'s
+reviewable, behaviour-preserving swap of `verify_wg_peer()`'s
 body; then land (2) as a follow-up commit for the read path
 (`wg get`, `link get`) where the sans-io surface is smallest,
 and keep (1) for the privileged mutating ops. Measure before
@@ -317,7 +306,7 @@ async def read_wg_peers(
 async def read_wg_pubkey(iface: str = 'wg0', ...) -> str: ...
 ```
 
-and `verify_wg_key()` becomes a thin composition over the two.
+and `verify_wg_peer()` becomes a thin composition over the two.
 Note the pure-getter rule: no `read_wg_peers(..., create=True)`.
 
 ---

ai/tpt-backends/README.md

@@ -7,9 +7,9 @@ different model/provider) without design or lib-selection drift.
 **Read [`00_shared_backend_contract.md`](./00_shared_backend_contract.md)
 first** — it is the normative description of what a `tractor`
 transport backend *is* as of `main@83b34884` (the backend
-duck-type, registration and address-selection wiring, the
+duck-type, the 10-item registration checklist, the test-harness
-test-harness plumbing, the code-style rules). The three plans
+plumbing, the code-style rules). The three plans assume it and
-assume it and document only their own deltas.
+document only their own deltas.
 
 | plan | issue | dep | size | lands |
 | --- | --- | --- | --- | --- |
@@ -23,12 +23,10 @@ Headline conclusions:
   `trio.SocketListener` are address-family agnostic (only
   `SOCK_STREAM` + a trio socket), and CPython ships `AF_TIPC` +
   23 `TIPC_*` constants. So the backend is ~one module of
-  contract boilerplate, zero new deps, and it buys kernel-native
+  contract boilerplate, zero new deps, and it buys
-  service primitives: `bind()` publishes, known-address
+  *kernel-native* service discovery: `bind()` publishes,
-  `connect()` resolves, and topology events report publication
+  `connect()`-by-name resolves — no registrar in the loop.
-  changes. Actor-name lookup, registrar state and split-brain-safe
+  (`modprobe tipc` is required; hard-gate everything.)
-  election remain separate work. (`modprobe tipc` is required;
-  hard-gate everything.)
 - **QUIC's cost is entirely in two adapters**, not in QUIC. The
   `iroh` python bindings are `uniffi`-generated asyncio, but the
   asyncio dependency is confined to *one* future-poll callback —

examples/multihost/wg_lan/README.md

@@ -49,9 +49,11 @@ uv sync
 ```
 
 gets you a `wg`-aware `multiaddr`. That pin goes away once a
-release carries the codec. Its `py-multibase` dependency provides
+release carries the codec. You also need `multibase`:
-the imported `multibase` module; no separate install command is
+
-needed.
+```bash
+uv pip install multibase
+```
 
 Without the codec `parse_wg_maddr()` raises immediately with an
 actionable message — there is deliberately **no** degraded
@@ -62,9 +64,8 @@ tunnel API (`.decapsulate_code()`, `.split()`, `.join()`,
 `.encapsulate()`, `.value_for_protocol()`) rather than any
 bespoke segment slicing — see its README "En/decapsulate" and
 "Tunneling" sections. gh #429 was about *dropping* our NIH
-parser, and that applies to peeling this composed maddr just as
+parser, and that applies to peeling a tunnel stack just as much
-much as to decoding one proto. This example rejects multiple
+as to decoding one proto.
-`/wg/` segments because `WGTunnelledAddr` stores one tunnel.
 
 ## 0. tunnel setup (out-of-band, both hosts)
 
@@ -105,11 +106,8 @@ AllowedIPs = 10.0.11.1/32
 PersistentKeepalive = 25
 ```
 
-This example configures host A's `ListenPort` and host B's
+Note how `ListenPort` and `Endpoint` are exactly the maddr's
-`Endpoint` from the maddr bearer, and configures host A's
+bearer segment, and `[Interface] Address` is its overlay host.
-`[Interface] Address` from its overlay host. The verification
-step below checks keys only; it does not inspect those fields or
-either peer's `AllowedIPs`.
 
 ```bash
 sudo wg-quick up wg0   # both hosts
@@ -120,40 +118,16 @@ ping -c1 10.0.11.1     # from B
 
 ```bash
 python -c "
-from wg_maddr import mb_pubkey
+import base64, multibase
 key = open('wg_pub.key').read().strip()
-print(mb_pubkey(key))
+print(multibase.encode('base64url', base64.b64decode(key)).decode())
 "
 ```
 
 Paste the `u...` output into `WG_MADDR` in both scripts (they use
 the same string — A's bearer, A's key, A's overlay ep).
 
-## 2. verify the keys
+## 2. run
-
-Interface inspection commonly needs `CAP_NET_ADMIN`. Keep that
-privileged operation separate from the `tractor` processes:
-
-```bash
-# host A: output must equal the maddr's A_pub key
-export WG_KEY_INSPECTION="$(sudo wg show wg0 public-key)"
-
-# host B: output must contain the maddr's A_pub key
-export WG_KEY_INSPECTION="$(sudo wg show wg0 peers)"
-```
-
-These checks establish only that host A uses the declared local
-key and host B has that key as a configured peer. They do not
-verify `Endpoint`, `AllowedIPs`, a recent handshake, or routing.
-The exported text contains public keys only. Each script passes it
-to `verify_wg_key()` with its host-specific role before starting
-`tractor`. Callers that already have permission to inspect the
-interface may omit that argument; the helper's direct query is
-async and requests cancellation after five seconds. Trio's
-subprocess termination escalation can make final process cleanup
-take longer than that cancellation deadline.
-
-## 3. run
 
 ```bash
 # host A
@@ -163,17 +137,6 @@ python host_a_srv.py
 python host_b_client.py
 ```
 
-Run both `tractor` programs as the normal application account,
-not as root. Privilege is needed only for tunnel setup and the
-separate inspection above. If using that preflight, keep the
-host-specific `WG_KEY_INSPECTION` value exported in each
-program's shell.
-
-The client binds its own actor listener to `10.0.11.2:0`, while
-the service actor binds to host A's `10.0.11.1` overlay host with
-a random port. Keep `LOCAL_OVERLAY_BIND` aligned with host B's
-WireGuard interface address if adapting this example.
-
 `host_a_srv.py` must be importable on host B too, since
 `portal.run()` refs the fn by module path — standard `tractor`
 RPC semantics.
@@ -190,19 +153,19 @@ Four corrections, all from
    all. `parse_wg_maddr()` now rejects it with an actionable
    error.
 2. **parsing is pure.** #482's helper had the key-check adjacent
-   to the parse; `verify_wg_key()` is now a separate, explicitly
+   to the parse; `verify_wg_peer()` is now a separate, explicitly
-   composed step for inspection-capable callers. A parser that
+   composed step that the caller invokes. A parser that shells
-   shells out is a nasty surprise.
+   out is a nasty surprise.
 3. **no `sudo`.** #482 ran `sudo wg show`; a library/example must
-   never escalate or run `tractor` as root. Privileged tunnel
+   never escalate. `wg show` works unprivileged for read on most
-   setup and key inspection are separate shell steps.
+   setups; if yours needs root, run the script as root rather
+   than embedding `sudo`.
 4. **no new `Address` proto-type.** The tunnel rides *beside* the
    overlay addr in a frozen `WGTunnelledAddr`, and only `.overlay`
    crosses into `open_nursery()`. #482 §6 floated a `WGAddress`
-   registered in `_address_types` — that registry maps available
+   registered in `_address_types` — that table is a `bidict`
-   transport keys to concrete address types, and
+   (1:1 proto-key↔type) and `_addr_to_transport` wants a
-   `_addr_to_transport` wants a `MsgTransport` per addr-type,
+   `MsgTransport` per addr-type, which `wg` doesn't have.
-   which `wg` doesn't have.
 
 ## next
 

examples/multihost/wg_lan/host_a_srv.py

@@ -6,17 +6,14 @@ Binds `tractor`'s registrar + an `echo_srv` sub-actor on the
 tunnel's *overlay* addr, declared as a single `wg` maddr.
 
 '''
-
 from __future__ import annotations
 
-import os
-
 import tractor
 import trio
 
 from wg_maddr import (
     parse_wg_maddr,
-    verify_wg_key,
+    verify_wg_peer,
     WGTunnelledAddr,
 )
 
@@ -37,17 +34,11 @@ async def echo(msg: str) -> str:
 
 async def main():
     addr: WGTunnelledAddr = parse_wg_maddr(WG_MADDR)
-    inspection: str | None = os.environ.get('WG_KEY_INSPECTION')
+    assert verify_wg_peer(addr), (
-    if not await verify_wg_key(
+        f'wg pubkey from maddr not active on wg0 !\n'
-        addr,
+        f'maddr: {WG_MADDR}\n'
-        role='local',
+        f'key: {addr.peer_pubkey}\n'
-        inspection=inspection,
+    )
-    ):
-        raise RuntimeError(
-            f'Maddr key is not wg0 local public key!\n'
-            f'maddr: {WG_MADDR}\n'
-            f'key: {addr.wg_pubkey}\n'
-        )
     print(
         f'wg bearer (kernel-owned): {addr.bearer}\n'
         f'tractor overlay ep: {addr.overlay}\n'
@@ -60,9 +51,7 @@ async def main():
     ) as an:
         await an.start_actor(
             'echo_srv',
-            bind_addrs=[(addr.overlay[0], 0)],
+            enable_modules=[__name__],
-            enable_transports=[addr.overlay_proto],
-            enable_modules=['host_a_srv'],
         )
         print(f'echo_srv up on\n  {addr.maddr}\n')
         await trio.sleep_forever()

examples/multihost/wg_lan/host_b_client.py

@@ -4,18 +4,15 @@ Host B: workstation dialing host A's actor tree through the
 `wg` tunnel.
 
 '''
-
 from __future__ import annotations
 
-import os
-
 import tractor
 import trio
 
 from host_a_srv import echo  # noqa: F401  (RPC refs it by mod path)
 from wg_maddr import (
     parse_wg_maddr,
-    verify_wg_key,
+    verify_wg_peer,
     WGTunnelledAddr,
 )
 
@@ -25,33 +22,23 @@ WG_MADDR: str = (
     '/wg/u<A_pub_b64url>'
     '/ip4/10.0.11.1/tcp/1616'
 )
-LOCAL_OVERLAY_BIND: tuple[str, int] = ('10.0.11.2', 0)
 
 
 async def main():
     addr: WGTunnelledAddr = parse_wg_maddr(WG_MADDR)
-    inspection: str | None = os.environ.get('WG_KEY_INSPECTION')
+    assert verify_wg_peer(addr), (
-    if not await verify_wg_key(
+        f'wg pubkey from maddr not a peer on wg0 !\n'
-        addr,
+        f'maddr: {WG_MADDR}\n'
-        role='peer',
+    )
-        inspection=inspection,
-    ):
-        raise RuntimeError(
-            f'Maddr key is not a configured wg0 peer!\n'
-            f'maddr: {WG_MADDR}\n'
-            f'key: {addr.wg_pubkey}\n'
-        )
     async with (
         tractor.open_root_actor(
             name='wg_client',
-            tpt_bind_addrs=[LOCAL_OVERLAY_BIND],
             registry_addrs=[addr.overlay],
             enable_transports=[addr.overlay_proto],
         ),
         tractor.find_actor(
             'echo_srv',
             registry_addrs=[addr.overlay],
-            raise_on_none=True,
         ) as portal,
     ):
         res: str = await portal.run(

examples/multihost/wg_lan/wg_maddr.py

@@ -26,7 +26,7 @@ stack "inner" reads as higher-up and later-called, whereas here
 the encapsulated addr is bound *first* and sits deeper in the
 maddr — two opposite intuitions on one word.
 
-`/wg/u<key>` itself carries a declared Curve25519 pubkey
+`/wg/u<key>` itself carries the tunnel peer's Curve25519 pubkey
 as multibase base64url (std base64 from `wg(8)` contains `/` and
 so can't go in a `/`-delimited maddr). It binds nothing at all;
 it's an identity, verified out-of-band.
@@ -43,33 +43,28 @@ Design rules this module follows (see
   through `.decapsulate_code()`, `.split()`, `.join()`,
   `.encapsulate()` and `.value_for_protocol()`. We hand-roll no
   segment splitting whatsoever — the whole point of gh #429 was
-  dropping the NIH parser, and that applies to peeling this
+  dropping the NIH parser, and that applies to *peeling a tunnel
-  composed maddr every bit as much as to decoding one proto.
+  stack* every bit as much as to decoding a single proto.
 - **parsing is pure**. `parse_wg_maddr()` does no I/O, no
   `subprocess`, no netlink. A parser that shells out is a nasty
   surprise.
 - **verification is an explicit, separate step**. The caller
-  composes `verify_wg_key()` when it has permission to inspect
+  composes `verify_wg_peer()` when it wants it; nothing implicit.
-  the iface; nothing implicit.
-- **exactly one `wg` segment is supported**. `WGTunnelledAddr`
-  stores one bearer and one key, so accepting another segment
-  would silently misrepresent the maddr.
 - **no new `Address` proto-type**. `wg` gets no entry in
-  `tractor.discovery._addr._address_types`, which maps available
+  `tractor.discovery._addr._address_types` (a `bidict`, so 1:1
-  transport keys to concrete address types, bc it has no
+  proto-key<->type) bc it has no `MsgTransport` of its own. The
-  `MsgTransport` of its own. The
   tunnel is a *bindspace*, so we carry it beside the overlay
   addr and strip to `.overlay` at bind/dial time.
 
 '''
-
 from __future__ import annotations
 import base64
+import subprocess
 from typing import Literal
 
 import msgspec
 from multiaddr import Multiaddr
-import trio
+from multiaddr.protocols import P_WG
 
 
 IPProto = Literal['ip4', 'ip6']
@@ -81,16 +76,15 @@ class WGTunnelledAddr(
 ):
     '''
     A `wg`-tunnelled endpoint: the underlay bearer, the tunnel
-    key, and the overlay addr `tractor` binds/dials.
+    peer key, and the overlay addr `tractor` binds/dials.
 
     '''
-
     # underlay, owned by `wg(8)`/the kernel — NEVER bound by us
     bearer: tuple[str, int]
 
-    # declared wg pubkey in std-base64 `wg(8)` form; it is the
+    # tunnel peer pubkey in the std-base64 `wg(8)` form, i.e.
-    # local key on the bearer host and a configured peer on a dialer
+    # directly comparable to `wg show <if> peers` output
-    wg_pubkey: str
+    peer_pubkey: str
 
     # overlay ep: an `UnwrappedAddress` as accepted by
     # `tractor.discovery.wrap_address()`
@@ -114,7 +108,7 @@ class WGTunnelledAddr(
         return (
             Multiaddr(f'/{self.bearer_ip}/{b_host}/udp/{b_port}')
             .encapsulate(
-                Multiaddr(f'/wg/{mb_pubkey(self.wg_pubkey)}')
+                Multiaddr(f'/wg/{mb_pubkey(self.peer_pubkey)}')
             )
             .encapsulate(
                 Multiaddr(
@@ -139,13 +133,7 @@ def mb_pubkey(wg8_key: str) -> str:
 
     '''
     import multibase
-
+    raw: bytes = base64.b64decode(wg8_key)
-    raw: bytes = base64.b64decode(wg8_key, validate=True)
-    if len(raw) != 32:
-        raise ValueError(
-            f'WireGuard public keys must decode to 32 bytes, '
-            f'not {len(raw)}'
-        )
     return multibase.encode('base64url', raw).decode('ascii')
 
 
@@ -155,17 +143,11 @@ def wg8_pubkey(mb_key: str) -> str:
 
     '''
     import multibase
-
     raw: bytes = multibase.decode(mb_key)
-    if len(raw) != 32:
-        raise ValueError(
-            f'WireGuard public keys must decode to 32 bytes, '
-            f'not {len(raw)}'
-        )
     return base64.b64encode(raw).decode('ascii')
 
 
-_wg_proto_known: bool | None = None
+_wg_proto_known: bool|None = None
 
 
 def _have_wg_maddr_proto() -> bool:
@@ -185,7 +167,6 @@ def _have_wg_maddr_proto() -> bool:
     if _wg_proto_known is None:
         from multiaddr.protocols import protocol_with_name
         from multiaddr.exceptions import ProtocolNotFoundError
-
         try:
             protocol_with_name('wg')
             _wg_proto_known = True
@@ -196,7 +177,7 @@ def _have_wg_maddr_proto() -> bool:
 
 
 def parse_wg_maddr(
-    maddr: str | Multiaddr,
+    maddr: str|Multiaddr,
 ) -> WGTunnelledAddr:
     '''
     Peel a `wg`-tunnelled maddr into its bearer/key/overlay
@@ -227,31 +208,18 @@ def parse_wg_maddr(
         for seg in segs
         for proto in seg.protocols()
     ]
-    wg_count: int = names.count('wg')
+    if 'wg' not in names:
-    if not wg_count:
         raise ValueError(
             f'Not a `wg`-tunnelled maddr, no `/wg/` segment ??\n'
             f'maddr: {ma}\n'
         )
-    if wg_count > 1:
-        raise ValueError(
-            f'Nested `wg` segments are not supported; '
-            f'`WGTunnelledAddr` stores one tunnel only.\n'
-            f'maddr: {ma}\n'
-        )
-
-    # Resolve the unreleased protocol only after the capability
-    # check, so importing this module works with released multiaddr.
-    from multiaddr.protocols import protocol_with_name
-
-    wg_code: int = protocol_with_name('wg').code
 
     # NOTE, `.decapsulate_code()` cuts at the LAST occurrence of
     # the proto and keeps the *prefix*, which is exactly the
     # bearer. It handles `/wg/` cleanly precisely bc it cuts on
     # proto-code and never tries to match an addr value — the
     # key seg has no addr of its own.
-    bearer_ma: Multiaddr = ma.decapsulate_code(wg_code)
+    bearer_ma: Multiaddr = ma.decapsulate_code(P_WG)
     overlay_ma: Multiaddr = Multiaddr.join(
         *segs[names.index('wg') + 1:]
     )
@@ -291,7 +259,7 @@ def parse_wg_maddr(
 
     return WGTunnelledAddr(
         bearer=bearer,
-        wg_pubkey=wg8_pubkey(ma.value_for_protocol('wg')),
+        peer_pubkey=wg8_pubkey(ma.value_for_protocol('wg')),
         overlay=overlay,
         overlay_proto=l4,
         bearer_ip=b_ip,
@@ -299,23 +267,13 @@ def parse_wg_maddr(
     )
 
 
-async def verify_wg_key(
+def verify_wg_peer(
     addr: WGTunnelledAddr,
-    role: Literal['local', 'peer'],
     iface: str = 'wg0',
-    timeout: float = 5,
-    inspection: str | None = None,
 ) -> bool:
     '''
-    Verify the declared key in the role required on this host.
+    True iff `addr.peer_pubkey` is a configured peer (or our own
-
+    pubkey) on `iface`.
-    A bearer host uses `role='local'`; a dialer uses `role='peer'`.
-    This verifies only key presence. It does not inspect the peer's
-    endpoint, AllowedIPs, handshake state, or iface addresses.
-
-    `inspection` accepts output captured by a separate privileged
-    `wg show` step. Without it, query asynchronously for callers
-    which already have interface-inspection permission.
 
     IMPURE + explicit by design: never called from
     `parse_wg_maddr()`.
@@ -326,23 +284,16 @@ async def verify_wg_key(
     netns unless `netns=` is passed down.
 
     '''
-    match role:
+    def _wg(*args: str) -> str:
-        case 'local':
+        return subprocess.run(
-            field = 'public-key'
+            ['wg', 'show', iface, *args],
-        case 'peer':
+            capture_output=True,
-            field = 'peers'
+            text=True,
-        case _:
+            check=True,
-            raise ValueError(f'Unknown WireGuard key role: {role!r}')
+        ).stdout
 
-    if inspection is None:
+    return (
-        with trio.fail_after(timeout):
+        addr.peer_pubkey in _wg('peers').split()
-            proc = await trio.run_process(
+        or
-                ['wg', 'show', iface, field],
+        addr.peer_pubkey == _wg('public-key').strip()
-                capture_stdout=True,
+    )
-                check=True,
-            )
-        inspection = proc.stdout.decode()
-
-    if role == 'local':
-        return addr.wg_pubkey == inspection.strip()
-    return addr.wg_pubkey in inspection.split()

uv.lock

@@ -273,9 +273,9 @@ name = "greenback"
 version = "1.2.1"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
-    { name = "greenlet" },
+    { name = "greenlet", marker = "python_full_version < '3.14'" },
-    { name = "outcome" },
+    { name = "outcome", marker = "python_full_version < '3.14'" },
-    { name = "sniffio" },
+    { name = "sniffio", marker = "python_full_version < '3.14'" },
 ]
 sdist = { url = "https://files.pythonhosted.org/packages/dc/c1/ab3a42c0f3ed56df9cd33de1539b3198d98c6ccbaf88a73d6be0b72d85e0/greenback-1.2.1.tar.gz", hash = "sha256:de3ca656885c03b96dab36079f3de74bb5ba061da9bfe3bb69dccc866ef95ea3", size = 42597, upload-time = "2024-02-20T21:23:13.239Z" }
 wheels = [
@@ -518,7 +518,7 @@ wheels = [
 [[package]]
 name = "multiaddr"
 version = "0.2.0"
-source = { git = "https://github.com/multiformats/py-multiaddr.git?rev=f86519daaa21699023d0037c58cdff600313dd09#f86519daaa21699023d0037c58cdff600313dd09" }
+source = { registry = "https://pypi.org/simple" }
 dependencies = [
     { name = "base58" },
     { name = "dnspython" },
@@ -533,6 +533,10 @@ dependencies = [
     { name = "trio-typing" },
     { name = "varint" },
 ]
+sdist = { url = "https://files.pythonhosted.org/packages/c7/10/4e26a8577cfce1c0febc8d83087e1373e93c695c6e73ad010546fb67e229/multiaddr-0.2.0.tar.gz", hash = "sha256:acb6b25c332ec1b2f1f8fef8d03a8c63385d34a87d690df0f4bba43cdf6efe8d", size = 58356, upload-time = "2026-03-17T21:51:00.274Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/b5/13/56e503d01218d1ca27ea9fda862045a4b400cae5e756f47315f5aaba0eee/multiaddr-0.2.0-py3-none-any.whl", hash = "sha256:bcff7bf3d7de3d6da0b865b25423bcb411de1d20d70cc6abfacf75170d17866c", size = 40424, upload-time = "2026-03-17T21:50:58.833Z" },
+]
 
 [[package]]
 name = "mypy-extensions"
@@ -1189,7 +1193,7 @@ requires-dist = [
     { name = "bidict", specifier = ">=0.23.1" },
     { name = "colorlog", specifier = ">=6.8.2,<7" },
     { name = "msgspec", specifier = ">=0.20.0" },
-    { name = "multiaddr", git = "https://github.com/multiformats/py-multiaddr.git?rev=f86519daaa21699023d0037c58cdff600313dd09" },
+    { name = "multiaddr", specifier = ">=0.2.0" },
     { name = "pdbp", specifier = ">=1.8.2,<2" },
     { name = "platformdirs", specifier = ">=4.4.0" },
     { name = "setproctitle", specifier = ">=1.3,<2" },