repo-docs
Skill0 starsChanged 15 days agoMITView on GitHub
Set up or reorganise a repository's user, developer and agent documentation.
16 other projects use a version of this.
Adopted by
Most recently taken up by
- korya/swd-skillstheir own version, first seen Nov 1, 2026
- certinia/debug-log-analyzer-mcptheir own version, first seen Oct 28, 2026
- ndthanhdev/aiotheir own version, first seen Oct 25, 2026
- bsene/skillstheir own version, first seen Oct 25, 2026
- rohanhong/tmpl.general-repotheir own version, first seen Oct 11, 2026
- OneDro1d/dark-factorytheir own version, first seen Oct 8, 2026
Other versions16 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 16
- YurunChen/repo-docs-skills★ 457same name
- certinia/debug-log-analyzer-mcp★ 36same name
- IndelibleVivi/repo-docs-curator★ 10same name
- phanghonghao/THU-Awesome-Skills★ 8same name
- vbonk/repo-template★ 8same name
- bsene/skills★ 7same name
- OneDro1d/dark-factory★ 4same name
- rohanhong/tmpl.general-repo★ 3same name
- korya/swd-skills★ 2same name
- IsthisLee/agent-context-manager★ 1same name
6 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository5 other files it gives its agents.
- librenms-os-supportSkill
- repo-workflowSkill
- setup-agent-skillsSkill
- unslopSkill
- writing-for-agentsSkill
The file
We split this skill into the rules it gives an agent
11 rules. 0 are common, 9 are distinctive, and 2 are very specific to this project.
0
Common
Most of the 13 edited versions we read keep these.
9
Distinctive
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 13
Read the existing README, agent instructions, developer and contribution guides, manifests, CI workflows and release configuration.
- Very specific: line 24
Use the following division, preserving existing useful paths such as docs/DEVELOPING.md rather than creating a competing guide:
- Distinctive: line 34
Create security, privacy, licence, support or release documents only when the project's actual needs justify them.
- Distinctive: line 38
Keep the README focused on users.
- Distinctive: line 39
Write runnable developer instructions from the current scripts and workflows.
- Distinctive: line 40
Document architectural reasons, ownership boundaries and compatibility traps that code alone does not explain.
- Distinctive: line 41
Keep agent instructions short.
- Very specific: line 43
Put the full commit/PR/issue standard in one repository-owned location and link to it from agent instructions.
- Distinctive: line 45
Use British spelling and synthetic examples.
- Distinctive: line 49
Every new factual statement is supported by inspected code, configuration, existing policy or a user decision.
- Distinctive: line 50
Every relative file link and heading anchor resolves, and the README, developer guide and agent entry points connect to their intended …
“Common” is measured: kept by at least 6 in 10 of the 13 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Repository documentation
Run only when the user invokes this skill. Create or reorganise documentation in the target repository, using its current code and configuration as evidence. Apply writing-for-agents when writing agent instructions. This task prepares documentation; commit, push, PR and publication actions follow the user's requested scope.
Inspect the repository
Read the existing README, agent instructions, developer and contribution guides, manifests, CI workflows and release configuration. Inspect the entry points and modules needed to establish the architecture. Identify:
- Who uses the project, what they receive and how they install or start it.
- How a developer builds, tests and runs it, including platform and credential prerequisites.
- Which repository owns shared code, how dependencies are consumed, and where platform boundaries sit.
- How changes, releases, support and security reports are handled.
Resolve contradictory documentation against current code and configuration. Ask for product, licensing or contribution-policy decisions that the repository cannot establish; mark unresolved facts explicitly rather than guessing. Continue when each proposed document has an audience and an evidence-backed purpose.
Give each fact one home
Use the following division, preserving existing useful paths such as docs/DEVELOPING.md rather than creating a competing guide:
| Document | Owns |
README.md | What the project does, who it is for, supported platforms, installation or a minimal quick start, and links to further help |
docs/DEVELOPMENT.md | Developer prerequisites, build/run/test steps, architecture boundaries, shared dependencies, local configuration/storage, and release or deployment procedures |
AGENTS.md / CLAUDE.md | A short agent entry point: orientation, high-impact invariants and conditional pointers to the developer or contribution guide |
| Existing contribution guide, or a section in the developer guide | Branch/PR/issue conventions, scope list, checks and merge policy |
| Release notes / What's New | User-visible changes for a release, separate from PR implementation notes |
Create security, privacy, licence, support or release documents only when the project's actual needs justify them. Preserve established policy and licence text. Keep roadmaps in the repository's existing issue/milestone system and link there rather than duplicating a changing feature inventory.
Write and connect
- Keep the README focused on users. Put build internals, source layout and developer release mechanics in the developer guide; link between them.
- Write runnable developer instructions from the current scripts and workflows. Explain prerequisites, placeholders, expected outcomes and non-obvious platform limitations. Link to authoritative configuration instead of copying complete dependency versions, workflow schedules or directory trees that are cheap to inspect.
- Document architectural reasons, ownership boundaries and compatibility traps that code alone does not explain. Distinguish local development builds, CI artefacts, previews and stable releases, including which actions actually publish.
- Keep agent instructions short. Inline invariants needed on every change; point to detailed material with the condition for reading it. For example: “Before changing shared API contracts, read the compatibility section in the developer guide.” Ensure each pointer names a real file or section.
- For both Codex and Claude support, prefer one canonical
AGENTS.mdwith a thinCLAUDE.mdthat explicitly directs Claude to read it. Preserve an established working arrangement if it already provides a single source of truth. Put repository-specific rules here; leave general workflow conventions inrepo-workflow, and make the repository's essential instructions usable by contributors who do not have that personal skill installed. - Put the full commit/PR/issue standard in one repository-owned location and link to it from agent instructions. Use the
repo-workflowconventions as defaults while preserving explicit repository policy, the actual default branch and relevant scopes.
Use British spelling and synthetic examples. Explain credential names and permissions without including values. Update links when consolidating material and preserve externally used paths or provide an appropriate redirect. Keep user-visible release notes separate from technical review notes.
Verify and hand back
- Every new factual statement is supported by inspected code, configuration, existing policy or a user decision; remaining uncertainties are explicit.
- Every relative file link and heading anchor resolves, and the README, developer guide and agent entry points connect to their intended targets.
- Build/test examples match current entry points. Run safe relevant examples or existing documentation checks where available; report examples requiring missing platforms, credentials or publication as unverified.
- Facts have one authoritative home; agent pointers state when their target should be read. Remove obsolete instructions and duplicated sections as part of the same edit.
- Inspect the final diff for accidental policy changes, customer data, credentials and unrelated edits. Report the documents changed and any unresolved decisions.
Basis
Uses DashyNMS's separation of user-facing READMEs, technical development guides and concise CLAUDE.md entry points. Product-specific build, storage, signing and release details belong in the target repository rather than this reusable skill.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

