Use portable Agent Skills metadata and one repo-local source for
both Claude Code and OpenCode.
Deats,
- replace harness-only frontmatter and dynamic context
- make `commit-msg` runtime-aware and review-first
- add a thin OpenCode command and integration docs
- guard `np.searchsorted()` guidance before OOB indexing
- record OpenCode prompt provenance
Prompt-IO: ai/prompt-io/opencode/20260721T205228Z_58ffe487_prompt_io.md
(this patch was generated in some part by `opencode` using `gpt-5.6-sol` (`openai`))
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
Gud Boi