Chalk Labs
Need help deploying update-docs?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/cloverink-update-docs
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: data
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/update-docs
What's inside
- Skill name
update-docs- Description
- Sync documentation with code changes. Auto-detects scope from git diff, updates tier 1-4 docs (core, architecture, journey, config), verifies stats. Phase D of /ship pipeline.
- License
- MIT
- Instructions
- 116 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
# /update-docs
Keep documentation in sync with code. Detects what changed, updates the right tiers, leaves changes staged for review.
## How it works
```
1. Detect changes → git diff vs the merge base with origin/main
2. Classify → categorize files into doc tiers
3. Update → edit relevant docs per tier
4. Verify stats → run stats tools, refresh counts
5. Stage → git add (no commit, the parent commits per phase)
```
## Bundling behavior
When called from `/ship` (Phase D):
- Apply doc updates
- Stage changes (`git add`)
- Do NOT commit; `/ship` commits as `docs(<scope>): sync documentation`
- Cover everything that ships: the original work **and** every A/R fix round
- Return summary of what was updated
When called standalone:
- Apply doc updates
- Commit as `docs: update documentation`
## Phase 1: Detect Changes
```bash
BRANCH=$(git branch --show-current)
BASE=$(git merge-base origin/main HEAD) || { echo "cannot resolve origin/main, git fetch first"; exit 1; } # never guess a base
git diff --name-only $BASE..HEAD
git diff --name-only HEAD # also include unstaged
```
<!-- CONFIGURE: Adjust exclude patterns for your project (e.g., lockfiles, generated files) -->
## Phase 2: Documentation Tiers
| Tier | Files | Trigger |
|------|-------|---------|
| 1 Core | `README.md`, `CLAUDE.md`, top-level package READMEs | Any significant change |
| 2 Architecture | `docs/architecture.md`, `docs/database-schema.md` | Structural changes |
| 3 Journey | `docs/journey/*.md` (user flows) | Business logic changes |
| 4 Claude Config | `.claude/rules/*.md`, `.claude/skills/**/references/*.md` | Tool/workflow changes |
<!-- CONFIGURE: Replace tier paths to match your project's doc layout. Remove tiers you don't use. -->
## Phase 3: Stats Verification
If your project has a stats sync tool, run it first: it's the single source of truth for counts in README/CLAUDE.md.
```bash
# Example: replace with your project's tool
bun run tools/stats-sync.ts --update # writes diffs into core docs
bun run tools/stats-sync.ts # check-only mode for CI
```
<!-- CONFIGURE: Replace with your stats sync command, or remove if you don't track counts in docs -->
If no automated tool, verify common counts manually:
```bash
# Components, endpoints, tests: adjust globs to your structure
find src/components -name "*.tsx" | wc -l
grep -rcE "@(Get|Post|Patch|Delete|Put)\(" --include="*.ts" src/api/
```
## Phase 4: Per-Change Update Checklist
| Change type | Update |
|-------------|--------|
| New feature | README features list, relevant journey doc, CLAUDE.md if it affects how AI agents work |
| New API endpoint | API doc / journey doc with endpoint table, CLAUDE.md endpoint count |
| New component | Component README, design system doc if it introduces a pattern |
| Database change | `docs/database-schema.md`, migration notes, related journey docs |
| Config change | Doc that explains the config option, rule files that reference it |
| Refactor | Update file paths in any doc that linked to moved/renamed files |
| Dependency upgrade | CHANGELOG, README if version is documented |
<!-- CONFIGURE: Add change types specific to your project -->
## Phase 5: Stage & Report
```bash
git add README.md CLAUDE.md docs/ .claude/
```
Report format:
```
Documentation Update Summary
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Scope: <auto-detected files>
Updated: path/to/doc.md: what changed
Verified: Components: X | Tests: X | Endpoints: X
Skipped: path/to/doc.md: already up to date
Manual: path/to/doc.md: needs human review (reason)
```
## Guidelines
- **Concise**: lead with the answer, not the context
- **Scannable**: tables, bullet lists, headers
- **Cross-reference, don't duplicate**: link to related docs
- **Match code reality**: every code example, path, count must match the current code
- **Stage everything, commit nothing** when called from `/ship`
## Exit conditions
- All affected tiers updated, OR
- Reported as "no doc-worthy changes" if diff is purely internal (formatting, comments, test-only)
- Manual-review items flagged with reasonFrom cloverink/shipwright, MIT.
More from cloverink/shipwright
- 0.524score
review-full
cloverink
Smart review entry point.
SkillBFreshness: BClaude Code +9 - 0.524score
ship
cloverink
One-command pipeline from "done coding" to "PR ready for review".
SkillBFreshness: BClaude Code +9 - 0.521score
parallel-builder
cloverink
Build multiple files/pages in parallel using subagents.
SkillBFreshness: BClaude Code +9 - 0.521score
review-ux-fix
cloverink
UX review → fix → re-review loop targeting 10/10.
SkillBFreshness: BClaude Code +9 - 0.521score
review-docs
cloverink
Verify Phase D doc updates are accurate.
SkillBFreshness: BClaude Code +9 - 0.521score
frontend-design
cloverink
Create distinctive, production-grade frontend interfaces.
SkillBFreshness: BClaude Code +9 - 0.521score
audit-full
cloverink
Branch-aware audit.
SkillBFreshness: BClaude Code +9 - 0.521score
skill-creator
cloverink
Meta-skill that teaches how to create effective Claude Code skills.
SkillBFreshness: BClaude Code +9 - 0.521score
push
cloverink
Smart commit + push + auto-create PR.
SkillBFreshness: BClaude Code +9 - 0.521score
review-code-fix
cloverink
Code review → fix → re-review loop until score ≥ 9.5.
SkillBFreshness: BClaude Code +9 - 0.518score
merge-pr
cloverink
Merge current PR after CI passes.
SkillBFreshness: BClaude Code +9
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Sep 24, 2026 (1 month ago)
- Last release
- Sep 24, 2026 (1 month ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/cloverink/shipwright
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from cloverink
- 0.524score
review-full
cloverink
Smart review entry point.
SkillBFreshness: BClaude Code +9 - 0.524score
ship
cloverink
One-command pipeline from "done coding" to "PR ready for review".
SkillBFreshness: BClaude Code +9 - 0.521score
parallel-builder
cloverink
Build multiple files/pages in parallel using subagents.
SkillBFreshness: BClaude Code +9 - 0.521score
review-ux-fix
cloverink
UX review → fix → re-review loop targeting 10/10.
SkillBFreshness: BClaude Code +9 - 0.521score
review-docs
cloverink
Verify Phase D doc updates are accurate.
SkillBFreshness: BClaude Code +9 - 0.521score
frontend-design
cloverink
Create distinctive, production-grade frontend interfaces.
SkillBFreshness: BClaude Code +9
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get update-docs running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
update-docs skill. Agent Mart. https://agentmart.dev/skills/cloverink-update-docs (last updated 2026-10-24)