Vercel Composition Patterns
supabase/supabase
React composition patterns that scale. An agent skill from supabase/supabase.
Research current code state then update corresponding docs, ensuring docs stay in sync with code.
Install Update Docs
$ npx skills add sd0xdev/sd0x-harness --skill update-docs -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install sd0xdev/sd0x-harness update-docs --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/sd0xdev/sd0x-harness.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/update-docs .claude/skills/update-docs && rm -rf skills-srcUse ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.
Claude Code skills documentation · loads skills from .claude/skills/
Install the "update-docs" agent skill from https://github.com/sd0xdev/sd0x-harness/tree/main/skills/update-docs into .claude/skills/update-docs/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "update-docs", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/sd0xdev/sd0x-harness/tree/main/skills/update-docsType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add sd0xdev/sd0x-harness --skill update-docs -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install sd0xdev/sd0x-harness update-docs --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/sd0xdev/sd0x-harness.git skills-src && mkdir -p .agents/skills && cp -r skills-src/skills/update-docs .agents/skills/update-docs && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "update-docs" agent skill from https://github.com/sd0xdev/sd0x-harness/tree/main/skills/update-docs into .agents/skills/update-docs/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "update-docs", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add sd0xdev/sd0x-harness --skill update-docs -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install sd0xdev/sd0x-harness update-docs --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/sd0xdev/sd0x-harness.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/skills/update-docs .cursor/skills/update-docs && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "update-docs" agent skill from https://github.com/sd0xdev/sd0x-harness/tree/main/skills/update-docs into .cursor/skills/update-docs/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "update-docs", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/sd0xdev/sd0x-harness.git --path skills/update-docs--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add sd0xdev/sd0x-harness --skill update-docs -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install sd0xdev/sd0x-harness update-docs --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/sd0xdev/sd0x-harness.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/skills/update-docs .gemini/skills/update-docs && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "update-docs" agent skill from https://github.com/sd0xdev/sd0x-harness/tree/main/skills/update-docs into .gemini/skills/update-docs/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "update-docs", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install sd0xdev/sd0x-harness update-docsInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add sd0xdev/sd0x-harness --skill update-docs -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/sd0xdev/sd0x-harness.git skills-src && mkdir -p .github/skills && cp -r skills-src/skills/update-docs .github/skills/update-docs && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "update-docs" agent skill from https://github.com/sd0xdev/sd0x-harness/tree/main/skills/update-docs into .github/skills/update-docs/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "update-docs", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add sd0xdev/sd0x-harness --skill update-docs -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install sd0xdev/sd0x-harness update-docs --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/sd0xdev/sd0x-harness.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/skills/update-docs .opencode/skills/update-docs && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "update-docs" agent skill from https://github.com/sd0xdev/sd0x-harness/tree/main/skills/update-docs into .opencode/skills/update-docs/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "update-docs", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Facts
- Skill name
update-docs- GitHub stars
- 192
- Token cost
- ~2.1k tokens
- SKILL.md length
- 1,093 words
- Files
- 1
- Skills in repo
- 89
- Repo updated
- First seen
- Licence
- MIT
At a glance
Research current code state then update corresponding docs, ensuring docs stay in sync with code.
- Works in 6 steps: Locate Docs and Related Code (5-Level… → 5: Classify Each Target — Sync… → Research Current Code State → …
- Development work in your project
- SKILL.md covers Trigger, When NOT to Use, Auto-Trigger and Task, plus 2 more sections
- Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md
What it does
Update Docs is an agent skill from sd0xdev/sd0x-harness. Research current code state then update corresponding docs, ensuring docs stay in sync with code.
Its SKILL.md is about 2.1k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.
It sits in Development. The repository describes itself as: The harness layer for Claude Code — a reference implementation of harness engineering with hook-enforced dual review, state-machine gates that survive context compaction, and… The licence is MIT.
When your agent uses it
- Development work in your project
Example prompts
- “/update-docs”
Requirements
- Pre-approved tools (allowed-tools): Read, Write, Edit, Grep, Glob, Bash(ls:*), Bash(git:*), Bash(find:*), Bash(node:*)
Workflow steps
6 steps, taken from the step headings in SKILL.md.
What it can do on your machine
Read from SKILL.md and the folder at commit a4d4bc1. It shows what the files ask for, not the result of running them.
Tool permissions
Pre-approves these tools, so the agent can use them without asking each time:
ReadWriteEditGrepGlobBash(ls:*)Bash(git:*)Bash(find:*)Bash(node:*)
From allowed-tools in the SKILL.md frontmatter.
Runs code
No scripts in the folder and no shell commands in SKILL.md (its code samples are markdown).
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md.
From URLs in SKILL.md, links to its own repository left out.
Credentials
Names no API keys, tokens, secrets or passwords.
From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.
Context cost
Update Docs loads about 2.1k tokens when it runs. Until then it costs about 27 tokens; SKILL.md has 1,093 words of instructions outside code blocks.
Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.
Safety
Auto-check passed
The automated check found no risky patterns in SKILL.md.
Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.
SKILL.md
The full file from sd0xdev/sd0x-harness at commit a4d4bc1, republished under its MIT licence (© sd0xdev). 1,093 words, ~2,139 tokens.
Download SKILL.mdSave it as
.claude/skills/update-docs/SKILL.md (or your agent's skills folder).- name
- update-docs
- description
- Research current code state then update corresponding docs, ensuring docs stay in sync with code.
- allowed-tools
- Read, Write, Edit, Grep, Glob, Bash(ls:*), Bash(git:*), Bash(find:*), Bash(node:*)
Update Docs
Trigger
- Keywords: update docs, sync docs, docs out of date, update-docs
When NOT to Use
- Document review (use
/codex-review-doc) - Creating new docs (use
/tech-specor/create-request) - Document refactoring (use
/doc-refactor)
Auto-Trigger
Auto-triggered after precommit Pass, only when the change maps to a feature under docs/features/ (see @rules/auto-loop.md § Tiers, gate sequence). Can also be invoked manually.
Task
Step 1: Locate Docs and Related Code (5-Level Cascade)
Key principle: can't find target → ## Gate: ⚠️ Need Human — don't guess or create new docs.
Use the shared feature context resolution algorithm (see @skills/create-request/references/feature-context-resolution.md):
scan_error gate. scan_error !== false ⇒ the source sets are unknown, not empty —
report it and take the ⚠️ Need Human exit rather than syncing against a corpus you could not enumerate — an unreadable
corpus and a feature with no documents both return empty, and the first one silently becomes
"nothing to sync". Gate on !== false, not
=== true: a {} payload from a shell fallback carries no such field at all, and a non-null key
is not evidence the sets are complete — scan_error rides alongside a resolved key.
| Confidence | Action |
|---|---|
| high/medium | Proceed with detected feature |
| low | Proceed with warning |
| null (not found) | Output ## Gate: ⚠️ Need Human — do not guess |
Step 1.5: Classify Each Target — Sync Authority, Freeze Records
This skill rewrites current-authority docs. It does not rewrite records.
resolveDocRole(path, source, taxonomy) in scripts/lib/doc-metadata.js answers which a file is,
and owesCodeAlignment(path, source, taxonomy) is the same answer as a boolean — a doc owes code
alignment exactly when its role is the fallback (current-authority) one.
Both take a repository path as the first argument — never a role label. Spelling it
owesCodeAlignment() invites the call that has already been made here once: passing the string
"Design record" where the path goes. That string matches no rule, so it falls through to
FALLBACK_ROLE and the function returns true — a fail-closed default reads exactly like an
affirmative answer, and the wrong reading was an instruction to rewrite a frozen record
(docs/features/push-gate-optin/review-log-push-gate-optin.md, round 43).
The four roles below are BUILTIN_ROLE_CONFIG.closed_set in that file, and the Examples column
states what its path_defaults patterns actually match — read it there, not from the phase number.
docs-numbering.md numbers documents by lifecycle phase; doc-metadata.js assigns authority
role. They are different axes, and reading the first as the second is what put 2-tech-spec.md in
the Current-authority row of this table until 2026-08-21 — an instruction to rewrite a frozen design
record, and the exact failure the paragraph below this table warns about.
| Role | Matches (path_defaults) | What this skill does |
|---|---|---|
| Current authority — owes code alignment | 4-implementation*; anything whose first segment is skills/rules/agents/commands; and the fallback, which is what README.md resolves through | Rewrite the sections the code changed |
| Design record — states a decision | Conventionally 0-feasibility-study*, 1-requirements*, 2-tech-spec*, 3-architecture* — but the pattern is ^[0-3]-(feasibility|requirements|tech-spec|architecture), a cross-product: any of those four prefixes with any of those four stems, sixteen names, not four (see below). Also intent-<feature>.md (the intent-records rule) — planner-written constraints the implementer checks work against | Do not rewrite. Append a dated > **Update(…)** note recording what later changed — except intent-*: touch it not at all, appends included. Amending intent is a human re-decision, never a sync; this skill only reports a tension between the code and an invariant or Non-goal, and the human edits the file or does not |
| Work record — states what was asked | anything under a requests/ segment | Do not rewrite — this skill does not touch it at all. /create-request --update may overwrite exactly four fields (Status, the Progress table, AC checkboxes, Progress.Note); everything else in the ticket is frozen, and a closed ticket is frozen entirely. That is the whole mutable set for an ordinary update. The single exception is a reported factual correction — a non-lifecycle recorded fact such as a path or a date, or an unfilled template placeholder that was never a statement about the ticket — which may also apply to a closed ticket, authorizes only that correction, and never unrelated trimming or cleanup; it must be stated in Progress.Note. It must never change Status, a Progress phase status, or AC checkbox state: those lifecycle fields stay governed by the freeze and the ordinary transition rules, or "the erroneous fact" becomes a label an agent can put on a lifecycle edit to walk past them. Both halves are defined once in skills/create-request/SKILL.md § Phase 4.5 |
| History record | review-log-*, adr-* | Do not rewrite. Append only |
Show full SKILL.md (376 more words)Show less
owesCodeAlignment(path, source, taxonomy) is resolveDocRole(path, source, taxonomy) === FALLBACK_ROLE — the Current authority role and nothing else, so the three record rows are one
decision, not three. Named by role, not by position: "the first row" was true only until somebody
reordered or inserted one, and a reordered table would then have silently redirected this
instruction at a record.
The four canonical pairings — 0-feasibility-study, 1-requirements, 2-tech-spec,
3-architecture — are a naming convention (@rules/docs-numbering.md), not what the classifier
tests. Measured 2026-08-21: 1-tech-spec.md, 3-requirements.md, 0-architecture.md and
2-feasibility-study.md all resolve to Design record, because the prefix and the stem are
matched independently. Listing only the four pairings invites the opposite reading — that an
off-convention name is a gap the classifier cannot see — and it is the reverse: the classifier is
wider than the convention, deliberately, so a mis-numbered spec is still protected from rewriting.
Where it does stop is the prefix: 4-tech-spec.md falls past [0-3] into the fallback and
resolves to Current authority, i.e. rewritable. A spec numbered outside the range loses the
protection its name suggests, so fix the number rather than relying on the stem.
A record that disagrees with today's code is not stale — that disagreement is the record. Editing
it to agree destroys the only copy of what was decided, and buys nothing: the reviewer reads records
under the record-diff profile, which carries no code-alignment obligation
(skills/doc-review/SKILL.md § Review Profiles).
Report a record you were pointed at rather than silently skipping it — "not in scope for rewriting" is a fact the caller needs, and silence reads as "synced".
Step 2: Research Current Code State
Key research items:
- Any new scripts / skills / commands added?
- Any modified logic in existing files?
- Any new configuration or rules added?
- Any API or interface changes?
Step 3: Compare Docs vs Code Differences
| Item | Doc Description | Current Code | Status |
|---|
Step 4: Update Docs
Update document content based on differences:
- Architecture diagrams (Mermaid sequenceDiagram / flowchart)
- Core service table
- API description
- Data model
Step 5: Verification
After update:
- Re-read updated document sections
- Verify all new modules are documented
- Verify all removed modules are cleaned up
Safety Valve
After doc sync, compare code diff against pre-sync baseline. If new code changes exist (e.g., lint:fix modified code), return to review loop.
Output
## Doc Update Report
| Document | Sections Updated | Status |
|----------|-----------------|--------|
## Changes Made
- <summary of each update>
## Verification
- [ ] New modules documented
- [ ] Removed modules cleaned
- [ ] Diagrams updated© sd0xdev, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
Files
Just SKILL.md in skills/update-docs of sd0xdev/sd0x-harness.
Open the folder on GitHubat commit a4d4bc1
Compare with similar skills
Update Docs next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Update Docs this skillsd0xdev/sd0x-harness | 192 | — | ~2.1k | Automated safety check: Pass | MIT | |
| Vercel Composition Patternssupabase/supabase | 111k | 59 repos | ~726 | Automated safety check: Pass | MIT | |
| Finishing a Development Branchobra/superpowers | 296k | 5 repos | ~1.9k | Automated safety check: Pass | MIT | |
| Typescript Advanced Typesrolling-scopes/rsschool-app | 10k | 25 repos | ~4.2k | Automated safety check: Pass | MPL-2.0 | |
| PR Babysitteropeninterpreter/openinterpreter | 69k | 3 repos | ~4.2k | Automated safety check: Pass | Apache-2.0 | |
| Code Review ChecklistshareAI-lab/learn-claude-code | 78k | 5 repos | ~1.1k | Automated safety check: Pass | MIT |
Similar skills
- Official
Finishing a Development Branch
obra/superpowers
Walks the last step of a branch: confirm tests pass, detect the git environment, ask how to integrate, carry out your choice and clean up the worktree.
Typescript Advanced Types
rolling-scopes/rsschool-app
Master TypeScript's advanced type system including generics, conditional types, mapped types, template literals, and utility types for building type-safe applications.
PR Babysitter
openinterpreter/openinterpreter
Watches an open GitHub pull request until it merges, handling review comments, diagnosing CI failures and retrying flaky checks along the way.
Code Review Checklist
shareAI-lab/learn-claude-code
Reviews code against a five-part checklist covering security, correctness, performance, maintainability and testing, and reports findings in a fixed format.
Greploop
onyx-dot-app/onyx
Iteratively improves a PR (GitHub), MR (GitLab), or shelved changelist (Perforce) until Greptile gives it a 5/5 confidence score with zero unresolved comments.
More from sd0xdev/sd0x-harness
All 89 skills in this repoAdr
sd0xdev/sd0x-harness
Write an Architecture Decision Record (ADR) for a feature — Context / Decision / Status / Consequences / Alternatives, filed as docs/features/<feature/adr-<NNN-<title.md with a 3-digit zero-padded…
Load PR Review
sd0xdev/sd0x-harness
Load GitHub PR review comments into AI session — analyze, triage, plan.
Next Step
sd0xdev/sd0x-harness
Change-aware next step advisor. An agent skill from sd0xdev/sd0x-harness.
Obsidian CLI
sd0xdev/sd0x-harness
Obsidian vault integration via official CLI. An agent skill from sd0xdev/sd0x-harness.
Orchestrate
sd0xdev/sd0x-harness
Agent-driven workflow orchestration (v1 report-only). An agent skill from sd0xdev/sd0x-harness.
PR Comment
sd0xdev/sd0x-harness
Post friendly review comments to a GitHub PR — prepare locally, preview, then submit as atomic review.
Related
Categories
Questions about Update Docs
What does Update Docs do?
Research current code state then update corresponding docs, ensuring docs stay in sync with code. Update Docs is an agent skill from sd0xdev/sd0x-harness. Research current code state then update corresponding docs, ensuring docs stay in sync with code.
When should I use Update Docs?
Update Docs fits situations like: development work in your project.
How do I install Update Docs in Claude Code?
Run `npx skills add sd0xdev/sd0x-harness --skill update-docs -a claude-code`. Or copy the skill folder (skills/update-docs in sd0xdev/sd0x-harness) into .claude/skills/update-docs in your project. Claude Code loads it when a task matches its description.
How do I install Update Docs in Codex?
Run `npx skills add sd0xdev/sd0x-harness --skill update-docs -a codex`. Or copy the skill folder (skills/update-docs in sd0xdev/sd0x-harness) into .agents/skills/update-docs in your project. Codex loads it when a task matches its description.
Can I use Update Docs in Cursor, Gemini CLI or GitHub Copilot?
Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add sd0xdev/sd0x-harness --skill update-docs -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/update-docs, .gemini/skills/update-docs, .github/skills/update-docs and .opencode/skills/update-docs in your project.
What does Update Docs need to run?
SKILL.md names no scripts, command-line tools or credentials: Update Docs is instructions for the agent only. Its frontmatter pre-approves these tools: Read, Write, Edit, Grep, Glob, Bash(ls:*), Bash(git:*), Bash(find:*), Bash(node:*).
Does Update Docs access the network?
SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.
Is Update Docs safe to install?
Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.
What licence does Update Docs use?
Update Docs is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.
How many tokens does Update Docs use?
About 2.1k tokens (SKILL.md is roughly 8.6k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.
What are the alternatives to Update Docs?
Skills that share tags, products or a category with Update Docs: Vercel Composition Patterns (supabase/supabase, 111k stars), Finishing a Development Branch (obra/superpowers, 296k stars), Typescript Advanced Types (rolling-scopes/rsschool-app, 10k stars) and PR Babysitter (openinterpreter/openinterpreter, 69k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains Update Docs?
sd0xdev (a GitHub user) maintains it in sd0xdev/sd0x-harness, which has 192 GitHub stars. The repository holds 89 skills in this directory. The repository was last updated on October 8, 2026.
Source: sd0xdev/sd0x-harness on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.