文档

Project Quality Governance

试用

Improves the quality of project code whilst boosting development efficiency; suitable for agents such as ChatGPT, Claude and OpenCode

它能做什么

Create, audit, or refactor authoritative technical specifications, issue and improvement logs, risk-based verification workflows, and automated documentation checks. Use when an engineering repository needs durable quality governance, structured history, task-routed documentation, or consistency validation. Do not use for ordinary product copy, unstructured personal notes, or documentation unrelated to engineering quality governance.

技能文档

Project Quality Governance

Build a governance system that is authoritative, searchable, minimally repetitive, and safe to evolve.

Workflow

  1. Inspect the repository before proposing changes. Locate existing contributor instructions, technical specifications, change logs, validation scripts, package commands, hooks, and ignored private areas.
  2. Establish the authority order: current user direction, invariant safety rules, verified code and tests, then historical plans.
  3. Make the smallest compatible structural change. Preserve useful headings, identifiers, links, and user history; correct facts instead of rewriting history.
  4. Put invariant rules and a task-to-section reading map at the top of the authoritative specification. Keep one H1 and a stable H2/H3 hierarchy.
  5. Separate current rules, verified implementation facts, historical plans, and private experiments.
  6. Maintain one status index for defects and improvements while giving each record type its own required fields. Append history, link superseded behavior, and never invent missing timestamps or evidence.
  7. Add a read-only validator for structure, identifiers, fields, evidence, and time consistency. Integrate it without weakening existing protections.
  8. Verify proportionally: L1 for low-risk documentation or presentation changes, L2 for one module or behavior, and L3 for a completed stage or commit boundary.
  9. Report the resulting structure, checks run, unresolved items, and next precise entry point.

Reference Routing

  • Read references/technical-spec-architecture.md for authoritative specification structure.
  • Read references/issue-and-improvement-log.md for schemas, indexes, templates, corrections, and release-note candidates.
  • Read references/verification-workflow.md for validation depth, automated checks, and commit preparation.
  • Use scripts/validate-quality-docs.mjs only after adapting its explicit configuration to the repository's schema.

Guardrails

  • Match the repository's language unless the user requests another language.
  • Do not overwrite contributor history or turn unknown facts into invented precision.
  • Do not introduce product rules, dependencies, services, or new scope merely to complete documentation.
  • Do not read private research or secrets unless the current task explicitly requires it and access is authorized.
  • Keep validators diagnostic-only; they must not rewrite documents.
  • Preserve existing pre-commit and security guards; add checks compositionally.

Completion Criteria

  • The authoritative document exposes invariant rules and task routing before details.
  • Heading levels are stable and navigable.
  • The status index and detailed records agree exactly.
  • Defects and improvements have distinct schemas, explicit evidence, and honest historical gaps.
  • Automated checks are reproducible and proportional verification has been performed.
  • No private research, secrets, build outputs, or repository-specific material has leaked into reusable artifacts.

相关技能

Senior Staff Engineer skill for AI coding agents. Enforces simplicity, reuse-first, minimal-diff changes, testability, and Clean Code discipline. Use this sk...

Agent skill quality checker. Input a skill directory or skill files; output trigger clarity, metadata issues, examples, safety boundaries, installability, po...

36 次安装

AI QA Agent — the final quality gate before delivery. Tests code for correctness, verifies data integrity, checks brand voice compliance, and reviews document formatting. Use this skill whenever the user needs to review, verify, check, audit, inspect, or validate any deliverable before it ships. Tri

1 次安装

Conducts multi-axis code review. Use before merging any change. Use when reviewing code written by yourself, another agent, or a human. Use when you need to assess code quality across multiple dimensi Use when 需要Development领域自动化处理、数据分析和流程编排时使用。不适用于无明确需求的模糊场景。

Audit another Codex skill for structural compliance, trigger quality, instruction clarity, reuse of scripts or references, and overall maintainability. Use w...

16 次安装

Check coding-model API quality, capability fit, and drift with LT-lite and B3IT-lite. Use when Codex needs to verify whether an OpenAI/OpenAI-compatible/Anth...

15 次安装