AGENT SKILL

wordcount-style

Review a diff for style drift in the wordcount codebase. Checks for pathlib vs os.path, bare except clauses, missing type hints on public functions, and print in library code. Use during PR review or after generating code that touches src/wordcount/.

License 许可
Apache-2.0
最近更新
2026-08-18
Stars
★ 10
Forks
2

01 前言 Front Matter

name
wordcount-style
description
Review a diff for style drift in the wordcount codebase. Checks for pathlib vs os.path, bare except clauses, missing type hints on public functions, and print in library code. Use during PR review or after generating code that touches src/wordcount/.

03 SKILL.md 正文

wordcount-style

Project-local skill. Reviews a diff against the conventions in CLAUDE.md and the constraints in HARNESS.md.

Checks

  • Any use of os.path → flag, suggest pathlib.Path equivalent.
  • Any bare except: (no exception type) → flag.
  • Any public function (top-level def name(...) in src/wordcount/) without a return-type annotation → flag.
  • Any print(...) in src/wordcount/ files (excluding the if __name__ == "__main__": block) → flag.
  • Any Union[X, Y] or Optional[X] annotation → flag, suggest X | Y or X | None.

Output

For each finding: file path, line number, the offending snippet, the suggested replacement, and the convention it violates (cite CLAUDE.md or the HARNESS.md constraint by name).

04 同仓库更多 Skills

techtalk
techtalk/ai-readiness-assessment

wordcount-style

Diff-time style review against CLAUDE.md and HARNESS.md. Checks pathlib vs os.path, bare except, type hints on public functions, print in library code, and old-syntax unions.

★ 10
techtalk
techtalk/ai-readiness-assessment

wordcount-style

Review a diff for style drift against CLAUDE.md and HARNESS.md. Checks pathlib vs os.path, bare except, type hints on public functions, print in library code, and old-syntax unions.

★ 10
techtalk
techtalk/ai-readiness-assessment

ai-readiness-rollup

Roll up existing AI readiness assessments across several repositories into one portfolio report. Use when the user asks to "roll up our assessments", "assess across repos", "portfolio view", "how do our repos compare", "multi-repo readiness", "compare our teams' habitats", or any request to summarise AI readiness across more than one subject. Reads the assessment-summary block from reports that already exist — it does not re-run assessments and asks no behavioural questions. Also surfaced via the `/ai-readiness-rollup` slash command.

★ 10
techtalk
techtalk/ai-readiness-assessment

ai-readiness-assessment

Run a self-contained AI literacy / AI readiness assessment for the current repository. Use when the user asks to "assess AI readiness", "run an AI readiness assessment", "check our AI literacy level", "where are we on the AI collaboration framework", "evaluate how we work with AI", "score our AI maturity", "check our habitat maturity", or any request to place the team on the Agentic Experience 5-Level Habitat Maturity Model (14 dimensions, L1–L5) and the Sovereign Engineer six-level cognitive ladder (L0–L5) across context engineering, architectural constraints, and guardrail design. Produces a timestamped assessment, a gap-anchored reading path through *The Sovereign Engineer*, and a single TechTalk engagement recommendation. Also surfaced via the `/ai-readiness-assess` slash command.

★ 10