ship
Skill42 starsChanged 4 months agoApache-2.0View on GitHub
session-indexer ship pipeline: issue → implement → review → merge → close. Usage: /ship [--yes|-y] [issue-number | title]
194 other projects use a version of this.
Adopted by
Most recently taken up by
- claudialnathan/skillstheir own version, first seen Nov 15, 2026
- saleem-hadad/skillstheir own version, first seen Nov 15, 2026
- yagizdo/quivertheir own version, first seen Nov 15, 2026
- felipeam86/pipe-skillstheir own version, first seen Nov 15, 2026
- ShipfoxHQ/shipfoxtheir own version, first seen Nov 15, 2026
- nightly-labs/openbottheir own version, first seen Nov 14, 2026
Used with
Projects with it also use ios-sync, open-gstack-browser and 10 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
Saqlain-811Skillios-sync
Regenerate the iOS debug bridge against the latest upstream gstack templates. (gstack)
Also in 6 projects that use this
seek888Skillopen-gstack-browser
Launch GStack Browser — AI-controlled Chromium with the sidebar extension baked in. Opens a visible browser window where you can watch every action in real time. The sidebar shows a live activity feed and chat. Anti-bot stealth built in. Use when asked to "open gstack browser", "launch browser", "connect chrome", "open chrome", "real browser", "launch chrome", "side panel", or "control my browser". Voice triggers (speech-to-text aliases): "show me the browser".
Also in 10 projects that use this
ayudaySkillsetup-deploy
Configure deployment settings for /land-and-deploy. Detects your deploy platform (Fly.io, Render, Vercel, Netlify, Heroku, GitHub Actions, custom), production URL, health check endpoints, and deploy status commands. Writes the configuration to CLAUDE.md so all future deploys are automatic. Use when: "setup deploy", "configure deployment", "set up land-and-deploy", "how do I deploy with gstack", "add deploy config".
Also in 19 projects that use this
ShreyRaviSkilllanding-report
Read-only queue dashboard for workspace-aware ship. Shows which VERSION slots are currently claimed by open PRs, which sibling Conductor workspaces have WIP work likely to ship soon, and what slot /ship would pick next. No mutations — just a snapshot. Use when asked to "landing report", "what's in the queue", "show me open PRs", or "which version do I claim next". (gstack)
Also in 9 projects that use this
zebbernSkilldebug-fix
Find and fix a bug or issue — from any source (GitHub issue, error message, user report, or observed behavior)
Also in 4 projects that use this
ramybarsoumSkilldocument-release
Post-ship documentation update. Reads all project docs, cross-references the diff, updates README/ARCHITECTURE/CONTRIBUTING/CLAUDE.md to match what shipped, polishes CHANGELOG voice, cleans up TODOS, and optionally bumps VERSION. Use when asked to "update the docs", "sync documentation", or "post-ship docs". Proactively suggest after a PR is merged or code is shipped.
Also in 23 projects that use this
Other versions194 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 194
- garrytan/gstack★ 142ksame name
- rtk-ai/rtk★ 87ksame name
- simstudioai/sim★ 30ksame name
- millionco/react-doctor★ 15ksame name
- nightly-labs/openbot★ 7.9ksame name
- BuilderIO/agent-native★ 7.7ksame name
- fallow-rs/fallow★ 5.3ksame name
- puncsky/system-design-and-architecture★ 3.6ksame name
- plugin87/ux-ui-agent-skills★ 3.2ksame name
- ShenSeanChen/waku-agent★ 2.2ksame name
184 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 repository12 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- apply-dreamingSkill
- debugSkill
- doubt-driven-developmentSkill
- find-bugsSkill
- fix-reviewSkill
- grill-meSkill
- housekeepingSkill
- improveSkill
- revivalSkill
- self-learnSkill
- session-endSkill
The file
Show its rules and text
We split this skill into the rules it gives an agent
16 rules. 0 are common, 6 are distinctive, and 10 are very specific to this project.
0
Common
Most of the 181 edited versions we read keep these.
6
Distinctive
This author's own choices. Worth a look.
10
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- could be automatic: a setting could enforce it
- Very specific: line 35
Flag detection: Check for --yes / -y anywhere in arguments.
- Distinctive: line 38
No argument → auto-select (see below)
- Very specific: line 53
Prefer issues in the active milestone.
- Very specific: line 55
If YES_MODE=false, present top candidate and wait:
- Very specific: line 62
If YES_MODE=true, print selection and proceed immediately.
- Distinctive: line 77
If issue is not open: stop — "Issue #{N} is already closed."
- Very specific: line 94
Before touching any code, scan the issue body and codebase for decisions that must be made first.
- Distinctive: line 105
Search codebase for 1–3 concrete options grounded in existing patterns
- Distinctive: line 148
Wait for completion. On error or no PR: print error, stop.
- Very specific: line 168
If every changed file matches *.md, *.txt, docs/, .github//*.md → skip /fix-review, go to STEP 3.
- Very specific: line 194
If checks pending: gh pr merge {PR_NUMBER} --repo valpere/session-indexer --auto --squash --delete-branch then poll every 30s (max 30 min).
- Very specific: line 197
If conflicts: stop — "PR #{PR_NUMBER} has merge conflicts.
- Distinctive: line 227
If the call fails: warn and continue.
- Very specific: line 258
If the log call fails or /self-learn isn't installed, don't block the pipeline on it — note it in the final report's Warnings and move on.
- Distinctive: line 294
Never force-push — rebase if branch diverged from main.
- Very specific: line 296
All review agents run inside code-generator — do not re-launch in /ship.
“Common” is measured: kept by at least 6 in 10 of the 181 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
Skill: /ship
Issue → Merged PR Pipeline
/ship → auto-select next open issue (asks confirmation)
/ship --yes → auto-select and proceed, skip UI prompts
/ship -y → same as --yes
/ship {issue-id} → ship a specific issue by number
/ship {issue title} → find by title, then ship
/ship --yes {issue-id} → skip UI prompts (technical ambiguities always ask)
→ resolve issue (fetch title + description)
→ analyze ambiguities (scan description + codebase → ask decisions before coding)
→ mark in-progress
→ code-generator (implement → tests → review agents → simplify → docs → PR)
→ /fix-review (multi-model rounds + Claude Arbiter — does NOT merge)
→ merge PR (/ship owns the merge)
→ post comment (summary: what changed, PR link)
→ close issue
→ log to /self-learn (if installed)
→ final reportOne command. Pauses only for genuine technical decisions before coding.
STEP 0: Resolve Issue
Flag detection: Check for --yes / -y anywhere in arguments. Set YES_MODE=true, strip flag.
Argument classification (after stripping flags):
- No argument → auto-select (see below)
- Numeric or
#N→gh issue view {N} --repo valpere/session-indexer --json number,title,body,url - Otherwise → title search
YES_MODE:
- Skips "Proceed?" confirmation prompts and title disambiguation
- Does NOT skip STEP 0.5 — technical decisions always require input
If no argument — auto-select
gh issue list --repo valpere/session-indexer --assignee @me --state open \
--json number,title,url,milestone,labels --limit 10Prefer issues in the active milestone. Within milestone, prefer priority:high / urgent labels.
If YES_MODE=false, present top candidate and wait:
Suggested: #{number} — {title}
Milestone: {milestone} | {url}
Proceed? [Y / pick another / cancel]If YES_MODE=true, print selection and proceed immediately.
If a title string was given
gh issue list --repo valpere/session-indexer --state open --search "{title}" \
--json number,title,url --limit 10One close match → use directly. Multiple → present up to 5. None → stop.
Validation
Extract ISSUE_NUMBER, ISSUE_TITLE, ISSUE_BODY, ISSUE_URL.
If issue is not open: stop — "Issue #{N} is already closed."
Print:
Shipping: #{ISSUE_NUMBER} {ISSUE_TITLE}
{ISSUE_URL}Mark in-progress:
gh issue edit {ISSUE_NUMBER} --repo valpere/session-indexer --add-label "in-progress"STEP 0.5: Analyze Task & Resolve Ambiguities
Before touching any code, scan the issue body and codebase for decisions that must be made first. Runs even in YES_MODE — technical decisions are not UI prompts.
Scan for:
- "Decision:", "Options:", "or (b)", "discuss before starting", "if/or" branches
- Two implementation strategies side-by-side
- Scope conditionals: "if not already fixed in #{N}" → verify in code
- Dependency on another issue's output
- Constants or limits where the value isn't specified
For each ambiguity:
- Search codebase for 1–3 concrete options grounded in existing patterns
- Classify: multiple options → present with tradeoffs; dependency → verify in code; no options → ask directly
- Collect all decisions in one pass before proceeding
Format:
Before I start coding, I need to resolve {N} decision(s):
1. {Ambiguity title}
Context: {one sentence}
A) {option} — {tradeoff}
B) {option} — {tradeoff}Zero ambiguities: print "No ambiguities — proceeding." and continue.
STEP 1: Launch Code Generator
Agent(code-generator):
"Implement: {ISSUE_TITLE}
Issue body:
{ISSUE_BODY}
Issue URL: {ISSUE_URL}
Implementation pipeline:
1. Git worktree for isolation
2. Implement the feature/fix
3. Run: go vet ./... && go test -race ./...
4. Launch test-generator for new hooks/utils/services — skip if changeset is
migrations, components, docs, or .claude config files only
5. Re-run go test -race ./... after test generation
6. Parallel review: static-analysis + security-reviewer
7. Apply review findings
8. code-simplifier → docs-maintainer
9. Create PR (include '{ISSUE_URL}' in body)
10. Return PR number + URL"Wait for completion. On error or no PR: print error, stop.
Extract PR_NUMBER, PR_URL, BRANCH_NAME.
Mark in-review:
gh issue edit {ISSUE_NUMBER} --repo valpere/session-indexer \
--add-label "in-review" --remove-label "in-progress"STEP 2: Run /fix-review
Docs-only check
gh pr diff {PR_NUMBER} --repo valpere/session-indexer --name-onlyIf every changed file matches *.md, *.txt, docs/**, .github/**/*.md →
skip /fix-review, go to STEP 3. Print:
Skipping /fix-review — PR is docs-only. Code Review Pyramid has nothing to evaluate.Mixed PRs (any code file present): /fix-review runs. When in doubt, run it.
Standard
/fix-review {PR_NUMBER}Multi-model rounds + Claude Arbiter. Does NOT merge — STEP 3 owns the merge.
Collect summary (fixed / skipped / open). If unresolved items remain, note them and continue.
STEP 3: Merge PR
gh pr merge {PR_NUMBER} --repo valpere/session-indexer --squash --delete-branchIf checks pending: gh pr merge {PR_NUMBER} --repo valpere/session-indexer --auto --squash --delete-branch
then poll every 30s (max 30 min).
If conflicts: stop — "PR #{PR_NUMBER} has merge conflicts. Resolve and re-run."
Timeout: 30 min. If still open: ask user to merge manually, then re-run /ship {ISSUE_NUMBER}
to complete the tracker update.
Verify:
gh pr view {PR_NUMBER} --repo valpere/session-indexer --json state,mergedAt --jq '{state,mergedAt}'STEP 4: Post Completion Comment
gh issue comment {ISSUE_NUMBER} --repo valpere/session-indexer --body "$(cat <<'EOF'
## Shipped — PR #{PR_NUMBER}
**Summary:** {one-paragraph description of what changed and why}
**Key changes:**
- {file or component}: {what changed and why}
**Tests:** {what was added or verified}
**PR:** {PR_URL}
EOF
)"If the call fails: warn and continue.
STEP 5: Close Issue
gh issue close {ISSUE_NUMBER} --repo valpere/session-indexer \
--comment "Shipped in PR #{PR_NUMBER}."On failure: warn, ask user to close {ISSUE_URL} manually.
STEP 6: Log to /self-learn (if installed)
Skip silently if .claude/skills/self-learn/ doesn't exist in this project.
/self-learn log's Step 1 is written for interactive human invocation
("Ask the user: what happened?") — /ship runs this step itself, without
a human in the loop, so do not wait for input. Classify and log
directly using what you already observed during this run, following the
self-learn LOG step's own field schema and "if the user describes the
event, classify it yourself" allowance:
- Smooth run, zero
/fix-reviewescalations → win /fix-reviewcaught something that should have been prevented earlier (wrong pattern, missed edge case, convention violation) → mistake- STEP 0.5 ambiguity analysis surfaced a decision that would otherwise have produced a wrong implementation → win
If the log call fails or /self-learn isn't installed, don't block the
pipeline on it — note it in the final report's Warnings and move on.
STEP 7: Final Report
## /ship complete — #{ISSUE_NUMBER} {ISSUE_TITLE}
Issue: {ISSUE_URL} (closed)
PR: {PR_URL} (merged, branch deleted)
/fix-review: {fixed} fixed · {skipped} skipped · {open} still open
Pipeline:
✓ Issue resolved + ambiguities cleared
✓ Code generated (code-generator)
✓ go vet ./... && go test -race ./... passed
✓ test-generator ran (or skipped — docs/migrations only)
✓ static-analysis | security-reviewer
✓ code-simplifier + docs-maintainer
✓ PR created + /fix-review (multi-model + Arbiter)
✓ PR merged (squash)
✓ Completion comment posted
✓ Issue closedAppend Warnings for any skipped items or non-fatal failures.
RULES
- Issue must be ready/approved before shipping — not auto-verified.
- One PR per issue — reuse existing branch PR if it exists.
- Never force-push — rebase if branch diverged from main.
/shipowns the merge (STEP 3), not/fix-review.- All review agents run inside code-generator — do not re-launch in
/ship. - Tracker updates are best-effort — surface failures, never silently skip.
- 30-minute merge timeout — stop and report. Never loop indefinitely.
- Docs-only PRs skip /fix-review — the Pyramid has nothing to evaluate.
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.

