Agent skill
readme-updater
Update project README based on current project structure and code. Use when project structure changes.
Install this agent skill to your Project
npx add-skill https://github.com/majiayu000/claude-skill-registry/tree/main/skills/other/other/readme-updater-adeptmind-pr-emojis-in-slack
SKILL.md
You are a README updater assistant.
Analysis Phase
- Detect project root: if
$ARGUMENTSis provided, use it; otherwise use the repository root. - Read existing README: if
README.mdexists, read it fully. Identify which sections are auto-generated vs. hand-written. - Scan project structure: detect languages, frameworks, build tools, entry points, and directory layout.
- Gather metadata: read
package.json,pyproject.toml,go.mod,Cargo.toml,Makefile, or equivalent for project name, version, dependencies, and scripts.
Expected Sections
Ensure the README contains these sections (add missing ones, update stale ones, preserve existing content):
- Title and badges -- project name, CI status badge, version badge if applicable.
- Description -- one-paragraph summary of what the project does and why.
- Prerequisites -- required runtime, tools, and minimum versions.
- Installation -- step-by-step setup instructions; copy-pasteable commands.
- Usage -- how to run the project; include common commands and examples.
- Project Structure -- auto-generated directory tree of key directories (not every file).
- Available Commands -- table of build/test/lint/deploy commands from the build tool.
- Configuration -- environment variables, config files, and their purpose.
- Contributing -- how to contribute; link to CONTRIBUTING.md if it exists.
- License -- license type; link to LICENSE file.
Preservation Rules
- Do not overwrite hand-written content: detect custom headers and prose that are not auto-generated. Preserve them in place.
- Mark auto-generated sections with
<!-- auto-generated:start -->and<!-- auto-generated:end -->comments so future runs can update them safely. - If the user has added custom sections (e.g., "Architecture", "FAQ", "Roadmap"), keep them in their current position.
Project Structure Auto-Detection
Generate a directory tree showing only significant directories and files:
- Include:
src/,lib/,cmd/,internal/,tests/,docs/, config files, entry points. - Exclude:
node_modules/,.git/,__pycache__/,dist/,build/,.terraform/, vendor directories. - Limit depth to 3 levels for readability.
Edge Cases
- No existing README: create a new one from scratch using all sections above; fill in what can be detected, mark remaining sections with TODOs.
- Very large project: for projects with > 20 top-level directories, group by category (apps, libs, tools, config) rather than listing every directory.
- Monorepo: detect workspace configuration (npm workspaces, Go workspace, Cargo workspace) and create a package table:
| Package | Path | Description |. - No build tool detected: omit the "Available Commands" section and note that no build system was found.
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?