sdd-docs
Skill0 starsChanged 37 days agoMITView on GitHub
Use when creating, reviewing, or maintaining professional module and API documentation, aligning README and standalone guides with an SDD project's supported implementation, checking documentation examples and links, or auditing documentation after code changes or across the project.
In this repository14 other files it gives its agents.
- sdd-conventionsSkill
- sdd-designSkill
- sdd-forgeSkill
- sdd-implementSkill
- sdd-integrate-featureSkill
- sdd-manageSkill
- sdd-orientSkill
- sdd-planSkill
- sdd-reportSkill
- sdd-specifySkill
- sdd-steerSkill
- sdd-tasksSkill
- sdd-tddSkill
- sdd-verifySkill
The file
Files
Maintain project documentation
Choose the requested scope: documentation for affected code, README or selected guides, or a project-wide audit. Read only the relevant references:
| Work | Load |
| Module docstrings, API documentation, and explanatory comments | in-code documentation |
| README, standalone guides, examples, and navigation | standalone documentation |
| Change-scoped or project-wide audit and amendment findings | review and findings |
Use a current sdd-orient handoff for applicable instructions, Git state, target paths, and ownership. sdd-manage coordinates the requested editing scope; the active implementation workflow, sdd-implement for main task work or sdd-steer for a checkpoint amendment, coordinates documentation maintenance after substantive code changes. Read project documentation policies, relevant layout, accepted design, SPEC and PLAN or their active feature equivalents, affected code and tests, and existing documentation as needed. Review can remain read-only; edits require an eligible Git worktree and established ownership of pending changes.
Follow the documentation style specified by governing project documents. Otherwise select an established professional style appropriate to the language and documentation tooling, such as Google-style Python docstrings. Keep documentation accurate, useful, and proportionate to the code's responsibility. Ensure every code module within the selected scope has a professional module docstring or its language's equivalent; a project-wide audit covers every project-owned code module. Report generated or externally maintained modules separately and follow project policy for editing them.
Maintain documentation against both accepted requirements and observed implementation. Clearly distinguish supported behavior from planned capabilities. If a discrepancy requires an amendment to SPEC, PLAN, design, layout, or another authoritative requirement, identify its location, issue, impact, and proposed amendment and report it to the user. Defer the decision to the user; do not amend the governing document or invoke its owning skill automatically. Continue independent documentation work and report anything dependent on that decision as unresolved.
Keep code edits confined to documentation unless additional work is authorized. Do not mark tasks complete, create commits, push, or change hosted objects here; the active implementation workflow owns those operations. For Markdown, separate every heading from adjacent content with blank lines; a heading at the beginning of a file or template needs no preceding blank line.
Return the reviewed scope, chosen style and source, documentation changed, checks and observed outcomes, unreviewed or excluded areas, and unresolved findings or user decisions. Use sdd-report for a completion summary when requested.
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.

