Distilled distributed-runtime + structured-concurrency
expertise for the `tractor` actor-tree, auto-applied (not
user-invocable) when working on daemon/service arch, RPC
eps, `to_asyncio` integration, cancellation semantics or
hang/wedge/skew debugging.
Deats,
- `SKILL.md`: the core mental model incl. the post-split
actor-tree taxonomy (`datad`/`brokerd`/`emsd`/etc.),
daemon lifecycle conventions, actor-local-state hazards
and `tractor` primitive usage as deployed here.
- `gotchas.md`: symptom -> cause -> fix entries distilled
from this branch's (datad|brokerd)-split debugging (eg.
un-warmed contract caches, stale IPC resps, double/bare
log records, ib client-id collisions).
- `debug-recipes.md`: actor-system forensics incl. wedged
actor triage, hang-proof test gating and regression vs
pre-existing attribution.
(this patch was generated in some part by [`claude-code`][claude-code-gh])
[claude-code-gh]: https://github.com/anthropics/claude-code
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Refactor `SKILL.md` to adhere to claude-code skills docs and
eliminate content duplication with the supporting file.
Deats,
- fix `allowed-tools` from YAML list to comma-separated string
per frontmatter spec.
* https://code.claude.com/docs/en/skills#frontmatter-reference
- drop ~220 lines of inlined style-guide content that duplicated
`style-guide-reference.md`; replace with compact "Quick
Reference" section + markdown link to the full guide.
- fix supporting file ref from backtick-code to proper markdown
link syntax: `[style-guide-reference.md](./...)`.
- inline the file-writing instructions (timestamp + hash filename
format) directly, replacing the now-broken `CLAUDE.md` ref.
Also,
- add missing "commit msg" footer variant to
`style-guide-reference.md` (previously only showed "patch").
- move `.claude/CLAUDE.md` -> `style-guide-reference.md` as
proper skill supporting file.
(this patch was generated in some part by [`claude-code`][claude-code-gh])
[claude-code-gh]: https://github.com/anthropics/claude-code
Reorganize all 5 skills from loose `.md` files (and one
partially-formatted `commit_msg/`) into the documented
`subdirectory/SKILL.md` format with YAML frontmatter.
Deats,
- `commit_msg/` -> `commit-msg/` w/ enhanced frontmatter:
`argument-hint`, `disable-model-invocation`,
`allowed-tools`, dynamic `!` context injection for
staged diff + recent log, `$ARGUMENTS` support
- `piker_profiling.md` -> `piker-profiling/SKILL.md` +
`patterns.md` for detailed profiling patterns
- `piker_slang_and_communication_style.md` ->
`piker-slang/SKILL.md` + `dictionary.md` +
`examples.md`
- `pyqtgraph_rendering_optimization.md` ->
`pyqtgraph-optimization/SKILL.md` + `examples.md`
- `timeseries_numpy_polars_optimization.md` ->
`timeseries-optimization/SKILL.md` +
`numpy-patterns.md` + `polars-patterns.md`
Also,
- all background skills use `user-invocable: false`
for auto-application when relevant.
- use a hyphen convention across all dir names.
- content is now split into supporting files linked from each
`SKILL.md`.
(this patch was generated in some part by
[`claude-code`][claude-code-gh])
[claude-code-gh]: https://github.com/anthropics/claude-code
Since apparently my commit-msg generator thingie stores the "training"
prompt content in this file by default.. REALLY this should be put into
a `SKILL.md` or similar later so that only truly global ctx content is
put here.
Gud Boi