Agent skill
readme-updater
Audit, create, or rewrite repository README.md files to enterprise-grade quality using a deterministic evidence workflow across purpose, architecture, setup, operations, security, governance, and quality gates. Use when users ask to update/standardize/harden README docs, fix inaccurate onboarding instructions, prepare production-ready repository documentation, or generate executive/ownership-ready README content.
Install this agent skill to your Project
npx add-skill https://github.com/majiayu000/claude-skill-registry/tree/main/skills/other/other/readme-updater
SKILL.md
Readme Updater
Overview
Apply a deterministic, non-speculative workflow to assess README quality, close enterprise gaps, and deliver a high-signal README that supports onboarding, operations, compliance, and maintainability.
Trigger Hints
Use this skill when prompts include terms such as:
- "update README", "rewrite README", "create README"
- "enterprise README", "production-ready docs", "onboarding docs"
- "standardize documentation", "harden docs", "fix inaccurate README"
Workflow
- Resolve target README scope and repository root.
- Collect repository evidence from source-of-truth files.
- Audit the current README against enterprise criteria.
- Define the target README structure and applicability.
- Rewrite README content using only verified evidence.
- Run verification gates and finalize with a gap report.
0) Resolve Target Scope (required)
- Default target is
README.mdat repository root. - If multiple README files exist, prioritize root README unless user explicitly requests another path.
- If the user-requested README path does not exist, create it only when the request implies creation; otherwise ask one precise clarification question before writing.
- If README is missing, create one using this skill's target structure.
- If the requested scope conflicts with repository evidence, keep user-requested scope and note constraints in the gap report.
1) Identify Repository Context
- Determine project type, runtime, package manager, delivery model, and ownership signals from repository files.
- Extract canonical facts from
package.json, lockfiles, build configs, infra manifests, workflow files, and deployment configs. - Build a short evidence map before writing using this format:
- Fact
- Source file path
- Section where the fact will appear
- Do not infer unsupported operational, security, or compliance claims.
2) Audit Enterprise Criteria
Score each area as complete, partial, or missing:
- Purpose and business context
- Architecture and boundaries
- Prerequisites and local setup
- Environment/configuration strategy
- Build, test, lint, and quality commands
- Runbook basics (run, debug, troubleshoot)
- Deployment and release process
- Observability/support ownership
- Security, secrets, and compliance notes
- Contribution and governance model
- Versioning/changelog references
Also record:
- Top 3 gaps that block onboarding or safe operation.
- Any claims in the current README that cannot be verified from repository evidence.
3) Define Target README Structure
Use this baseline structure, then tailor to repo realities:
- Title and one-sentence value proposition
- Overview and scope
- Architecture at a glance
- Prerequisites
- Quick start
- Configuration and environment variables
- Development workflow (test/lint/build)
- Operations and troubleshooting
- Deployment and release
- Security and compliance
- Contributing and code standards
- Ownership/support and escalation
- License and legal notices
If sections are inapplicable, keep the heading and explicitly mark Not applicable with brief rationale.
4) Rewrite Rules
- Prefer precise, executable commands over prose.
- Keep language neutral, concise, and professional.
- Convert ambiguous wording into verifiable statements.
- Use tables for env vars, ports, and command catalogs.
- For TODO or unknown values, mark
TBD (owner needed)rather than guessing. - Preserve repository-specific terminology and established naming.
- Keep heading hierarchy valid and consistent for Markdown renderers.
- Prefer relative repository paths for file references and local workflows.
5) Verification Gates
Before finalizing:
- Validate all commands against repository tooling.
- Confirm referenced files/paths exist.
- Ensure section order and headings are consistent.
- Check that setup instructions are minimal and reproducible.
- Remove stale/duplicated instructions.
- Confirm links are valid repository-relative links or resolvable absolute URLs.
- Verify no unverifiable claims remain without
TBD (owner needed)marker.
Minimum verification output:
- Command validation result (
pass/fail) with corrected replacements if failures were found. - Path/link validation result (
pass/fail) with removed or corrected targets. - Claim verification result identifying corrected unverifiable statements.
6) Output Contract
When applying this skill, return:
- Updated
README.md. - A compact gap report listing unresolved
TBDitems. - A short verification summary listing what was validated (commands, paths, links).
- Optional follow-up tasks only when they unblock README completeness.
Reference
Load references/enterprise-readme-checklist.md when you need detailed review prompts and section-level acceptance criteria.
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
agent-ops-spec
Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.
agent-ops-state
Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.
agent-ops-spec
Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.
agent-ops-testing
Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.
agent-ops-testing
Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.
agent-ops-state
Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.
Didn't find tool you were looking for?