Update Docs logo

Update Docs

Organization
getnao
update-docs

Detect user-facing features in recently merged PRs and open a PR in the docs repo adding the missing documentation.

Overview

Publishergetnao
Repositorysylph
Skill nameupdate-docs
Stars
216
Forks
62
Bundled files
Instructions only
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by getnao on GitHub. Read the source before you install it.

Installation

Install the Update Docs AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/getnao/sylph/tree/main/.claude/skills/update-docs
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/getnao/sylph.git /tmp/sylph
mkdir -p .claude/skills
cp -r /tmp/sylph/.claude/skills/update-docs .claude/skills/update-docs
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Update Docs in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Update Docs on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Update Docs is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Update Docs

When the user runs /update-docs, scan recently merged PRs for user-facing changes and open a documentation PR.

MCP connectors

ConnectorPurpose
GitHubRead merged PRs, create docs PRs

Prerequisites

Both repos must be cloned locally:

  • Product repo: the main codebase (configured in CONTEXT.md)
  • Docs repo: the documentation site (configured in CONTEXT.md or asked on first run)

State file: agents/cmo/_state/docs_last_refresh.txt (plain text, single ISO date).


Step 1: Determine the PR window

  1. Read agents/cmo/_state/docs_last_refresh.txt if it exists
  2. If the file exists, use its date as SINCE
  3. If it does NOT exist (first run), use today - 15 days as SINCE
  4. Record TODAY as the date to write back at the end

Step 2: List merged PRs

bash
gh pr list -R <product-repo> \
  --state merged \
  --search "merged:>=<SINCE>" \
  --limit 100 \
  --json number,title,url,mergedAt,body,author,files

Keep the full list - every PR appears in the final summary, even those that don't need docs.


Step 3: Classify each PR

For each PR, decide: user-facing or not?

User-facing (needs docs):

  • New feature in the UI, CLI, API, or configuration
  • New command, flag, setting, integration, or UI element
  • Behavior change that affects how users interact with the product
  • New capability that a user would want to know about
  • Anything that belongs in release notes

NOT user-facing (skip):

  • Refactors, internal cleanups, tests, CI, infra
  • Dependency bumps, lockfile updates
  • Backend-only changes invisible to users
  • Bug fixes that restore already-documented behavior
  • Brand, marketing, or internal tooling changes

When unsure, read the PR body and changed files. If still ambiguous, err on no docs and note it in the summary.


Step 4: Fetch current docs state

Pull latest from the docs repo:

bash
cd <docs-repo> && git checkout main && git pull

Before writing any new doc, search the repo for existing coverage (grep on keywords from the PR title/body). If a relevant page exists, update it instead of creating a new one.


Step 5: Write the docs

For each user-facing PR that needs docs:

  1. Identify the right section and subfolder
  2. Either update an existing page or create a new one
  3. Follow the style of neighboring pages - tone, frontmatter, headings, code blocks
  4. If you create a new page, add it to the navigation/sitemap config
  5. Keep docs factual and concise - no marketing voice, direct tone
  6. Only document what actually shipped - read the PR diff if the description is thin. Never invent behavior.

Step 6: Create a new branch and PR

bash
cd <docs-repo>
git checkout -b docs/refresh-<TODAY>
git add <changed files>
git commit -m "docs: refresh for PRs merged since <SINCE>"
git push -u origin docs/refresh-<TODAY>
gh pr create -R <docs-repo> \
  --title "docs: refresh for PRs merged since <SINCE>" \
  --body "<PR summary - see template below>"

PR summary template

Every PR from Step 2 must appear in the list, in merge order. Use these exact shapes:

- PR merged #123 <PR title> - no docs added
- PR merged #124 <PR title> - added docs - <one-line description of what was documented and where>

No other bullet shapes. No section grouping. Keep it a flat list.

If zero PRs were user-facing, skip PR creation and just update the state file.


Step 7: Log the refresh

Write TODAY (ISO date) to agents/cmo/_state/docs_last_refresh.txt:

bash
echo "<TODAY>" > agents/cmo/_state/docs_last_refresh.txt
git add agents/cmo/_state/docs_last_refresh.txt
git commit -m "chore(cmo): log docs refresh <TODAY>"
git push

The next run picks up from this date.


Step 8: Report back

Return to the user:

  • Docs PR URL (or "no user-facing PRs this run" if none)
  • Number of PRs reviewed / number that got docs
  • The full PR summary list (same content as the PR body)
  • New last-refresh date

Guardrails

  • Never modify source code - docs changes only
  • Link to source PRs in the docs PR description
  • One docs PR per batch - group related changes together
  • Flag ambiguity - if a PR's user impact is unclear, ask before documenting
  • Never invent features - only document what the diff shows

Self-improvement

After the docs PR is reviewed and merged:

  1. If the reviewer requests changes to doc structure, tone, or coverage scope, update the relevant step in this skill file
  2. If a PR classification was wrong (documented something that didn't need docs, or missed something that did), refine the classification guide in Step 3
  3. Update _insights.md with patterns: which PRs consistently need docs, which doc styles get approved without changes
  4. If the PR summary format needs adjustment based on reviewer feedback, update the template in Step 6

Frequently asked questions

What does the Update Docs AI skill do?

Detect user-facing features in recently merged PRs and open a PR in the docs repo adding the missing documentation.

Why use Update Docs on TypingMind?

Because you install it once and use it with any model. Update Docs is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Update Docs in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/getnao/sylph/tree/main/.claude/skills/update-docs. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Update Docs?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Update Docs?

As many as you like. As long as a model supports skills, you can use Update Docs with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Update Docs AI skill free?

It is published on GitHub by getnao. Check the repository for licensing terms. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇