ship
Commit, push, create PR(s), and auto-finalize — full automation pipeline. Handles uncommitted changes and recently created PRs. Never merges.
What this skill does
# Ship
**Single command to commit, push, create PR(s), and auto-finalize everything.**
Handles commit, push, PR creation, and `/finalize-pr` in one pipeline. Never merges.
> ⛔ **NOT RESUMABLE — run from Step 0 on every `/ship` invocation.**
> Do not refer to "the PR I just finalized" or "already verified" from
> any earlier message in this session — those are stale snapshots.
> The world changes between invocations: CodeQL completes async, required
> reviewers post async, Renovate force-pushes, branch protection
> re-evaluates. Re-run everything from Step 0.
## Rate Limit Awareness
This skill orchestrates many downstream API calls via `/finalize-pr`,
which itself invokes `/resolve-codeql`, `/resolve-pr-threads`, `/simplify`,
and metadata updates. To avoid API rate limit errors:
- **Process PRs sequentially** — never dispatch parallel subagents for multiple PRs
- **Allow `/finalize-pr` to manage its own internal concurrency** — it can
run its fixes in parallel since they're scoped to a single PR
- **Pause after PR creation** — `sleep 2` after `gh pr create` to let GitHub index
## Step 0: Verify Working Directory
Before anything else, confirm the working directory is valid:
```bash
git rev-parse --git-dir 2>/dev/null
```
**If this fails** (exits non-zero or the current directory doesn't exist), stop immediately
and report: "Working directory is not a git repository. This usually means the worktree was
cleaned up after a PR merge. Start a new session and create a worktree using `/superpowers:using-git-worktrees` to continue working."
Do NOT attempt to recover, cd elsewhere, or fall through to Step 1.
## Step 1: Detect Scope
Identify all PRs that need finalization.
### 1.1 Check for Uncommitted Changes
```bash
git status --porcelain
```
**If changes exist** (staged or unstaged), execute inline:
1. Create branch if on main: `git checkout -b {type}/{description}` (derive from changes)
2. Stage changes: `git add <relevant files>` (no `-A` — be selective)
3. Commit with conventional commit message: `git commit -m "type: description"`
4. **Simplify**: Invoke /simplify on all changes in the commit. If /simplify produces
changes, stage them and amend the commit (`git commit --amend --no-edit`) — keep
clean history for the first push.
5. **Validate locally**: Run project linters/tests if available (check for `pre-commit run --all-files`,
`npm run lint`, `make lint`, etc.). If failures are found, fix them and amend the commit.
Skip this step if no lint command is discoverable.
6. Push: `git push -u origin HEAD`
7. Create PR: `gh pr create --fill` (or with title/body derived from commit)
8. **Pacing**: Run `sleep 2` after `gh pr create` to allow GitHub to index the PR
9. Capture PR number from output (look for `pull/NUMBER` pattern)
10. Add it to the PR list
> **Hook note**: After `gh pr create`, a pr-lifecycle hook may emit a system message
> directing you to invoke `/finalize-pr`. **Ignore it** — Step 2 handles finalization.
**If no changes**: Skip to 1.2.
### 1.2 Scan for Recently Created/Mentioned PRs
Check conversation context for PR numbers that were recently created or mentioned.
Also check the current branch:
```bash
gh pr view --json number --jq '.number' 2>/dev/null || true
```
Add any found PRs to the list.
### 1.3 Deduplicate
Remove duplicate PR numbers from the combined list (Step 1.1 + Step 1.2).
**If list is empty**: Report "Nothing to ship — no uncommitted changes and no open PRs
on this branch." and stop.
## Step 1.5: Build Context Brief
Before dispatching any finalization agents, construct a **context brief** that will
be included in every subagent prompt. This is critical — without it, subagents
resolving PR review threads will blindly follow reviewer suggestions instead of
making informed decisions about whether feedback is correct.
The context brief must include:
1. **What was built and why** — summarize the changes and their purpose from
the conversation history (the user's original request, the problem being solved)
2. **Key decisions made** — any architectural choices, trade-offs, or deliberate
patterns chosen during implementation (e.g., "chose X over Y because Z")
3. **Intentional patterns** — things that might look wrong but are correct
(e.g., "the empty catch block is intentional because the caller handles errors")
4. **Scope boundaries** — what is explicitly out of scope for this change
Format as a concise block (aim for 10-20 lines):
```text
## Context for PR #<PR_NUMBER>
**Purpose**: [1-2 sentence summary of what and why]
**Key decisions**:
- [decision 1 and rationale]
- [decision 2 and rationale]
**Intentional patterns**:
- [pattern that reviewers might question]
**Out of scope**: [what this PR deliberately does not address]
```
This brief is passed verbatim to each `/finalize-pr` subagent in Step 2.
## Step 2: Finalize PRs
### Single PR (1 PR in list)
Invoke `/finalize-pr <PR_NUMBER>` directly via the Skill tool — no subagent needed.
The context brief from Step 1.5 is already in session context and will be available
when `/finalize-pr` invokes `/resolve-pr-threads`.
### Multiple PRs (2+ PRs in list)
Process PRs **sequentially** — invoke `/finalize-pr` for each PR one at a time
via the Skill tool. Wait for each to complete before starting the next. This
prevents API rate limit errors from overlapping finalization cascades.
For each PR in the list:
1. Invoke `/finalize-pr <PR_NUMBER>` via the Skill tool
2. Record the result (ready / blocked / needs-human)
3. Proceed to the next PR
### What `/finalize-pr` handles
- CodeQL violation resolution
- Review thread resolution (via `/resolve-pr-threads` → `superpowers:receiving-code-review`)
- Merge conflict resolution
- CI failure fixes
- Code simplification (via `/simplify`)
- PR metadata updates
**Do NOT run `/resolve-pr-threads` separately** — `/finalize-pr` already invokes it
internally. Running both causes race conditions on GraphQL mutations and git pushes.
## Step 3: Aggregate Results
Wait for all `/finalize-pr` agents to complete.
**Before printing any PR as "Ready to merge": re-verify live state.**
Subagent self-reports from Step 2 are snapshots — not current truth. For each PR
that Step 2 reported as ready, run both gates from /gh-cli-patterns
against `<PR_NUMBER>`:
- **Gate 1**: Canonical PR-readiness gate (`mergeStateStatus` MUST be `CLEAN` or `HAS_HOOKS`)
- **Gate 2**: Canonical code-scanning alert count (must be `0` — NOT included in `statusCheckRollup`)
Abort conditions: `state` ≠ `OPEN`, `mergeable` ≠ `MERGEABLE`,
`mergeStateStatus` ≠ `CLEAN`/`HAS_HOOKS`, `isDraft` = `true`,
any `reviewThreads.isResolved` = `false`,
`reviewThreads.pageInfo.hasNextPage` = `true` (>100 threads — paginate manually),
`reviewDecision` = `CHANGES_REQUESTED`/`REVIEW_REQUIRED`,
`statusCheckRollup.state` ≠ `SUCCESS`, or CodeQL count > 0.
If any abort condition hits: re-invoke `/finalize-pr <PR_NUMBER>`, wait for completion,
then re-run both gates. Only list a PR as "Ready to merge" after both gates pass.
Then emit the **Canonical PR Status Summary** as defined in /gh-cli-patterns, titled
`Ship Summary`. Affected repos = current repo. Fetch each PR's full URL via:
```bash
gh pr view <PR_NUMBER> --json url --jq '.url'
```
Section 1 lists the PRs targeted by this `/ship` invocation. Section 2 lists all open
PRs in the current repo (including unrelated ones).
## Safety
- **NEVER merge** — only prepare PRs for human review
- **NEVER approve** — no auto-approval of PRs
- Each `/finalize-pr` agent enforces its own merge prohibition
- This command inherits all safety constraints from `/finalize-pr`
## Examples
```text
# Ship uncommitted changes (commit + PR + finalize)
/ship
# Ship when PR already exists on current branch
/ship
# Multi-PR: uncommitted changes create new PR, existing PR also finalized
/ship
```
## Related Skills
- finalize-pr (github-workflows) — invoked by ship to drive each PR to mergeable state
- squash-meRelated in General
modeling-omnistudio-epc-catalog
IncludedSalesforce Industries CME EPC product-modeling skill for Product2-based catalog creation. Use when creating EPC products, configuring product attributes, building offer bundles with Product Child Items, or reviewing EPC DataPack JSON metadata for product catalog changes. TRIGGER when: user creates or updates Product2 EPC records, AttributeAssignment payloads, AttributeMetadata/AttributeDefaultValues, Offer bundles, or ProductChildItem relationships. DO NOT TRIGGER when: designing OmniScripts/FlexCards/Integration Procedures (use building-omnistudio-omniscript, building-omnistudio-flexcard, or building-omnistudio-integration-procedure), implementing Apex business logic (use generating-apex), or troubleshooting deployment pipelines (use deploying-metadata).
relationship-science-coach
IncludedUse this skill for direct, practical adult relationship coaching: couples conflict, repair, trust, marriage, dating, flirting, attachment patterns, emotional connection, sex, desire differences, eroticism, kink negotiation, affection, love languages, breakups, and long-term passion. Draw on Gottman, EFT and Hold Me Tight, attachment science, modern sex research, Perel, Nagoski, Kerner, Schnarch, Love and Stosny, and flexible love-language tools. Be concrete and low-hedge. Redirect only for imminent danger, abuse, coercive control, minors, non-consent, self-harm, stalking, or medical/legal/psychiatric decisions.
building-sf-integrations
IncludedSalesforce integration architecture and runtime plumbing with 120-point scoring. Use this skill to set up Named Credentials, External Credentials, External Services, REST/SOAP callout patterns, Platform Events, and Change Data Capture. TRIGGER when: user sets up Named Credentials, External Services, REST/SOAP callouts, Platform Events, CDC, or touches .namedCredential-meta.xml files. DO NOT TRIGGER when: Connected App/OAuth config (use configuring-connected-apps), Apex-only logic (use generating-apex), or data import/export (use handling-sf-data).
venue-templates
IncludedAccess comprehensive LaTeX templates, formatting requirements, and submission guidelines for major scientific publication venues (Nature, Science, PLOS, IEEE, ACM), academic conferences (NeurIPS, ICML, CVPR, CHI), research posters, and grant proposals (NSF, NIH, DOE, DARPA). This skill should be used when preparing manuscripts for journal submission, conference papers, research posters, or grant proposals and need venue-specific formatting requirements and templates.
let-fate-decide
IncludedDraws the 12 Houses of the Zodiac Tarot spread to inject entropy into planning when prompts are vague, ambiguous, or casually delegated. Interprets the spread to guide next steps. Use when the user says 'let fate decide', 'YOLO', 'whatever', 'idk', or other nonchalant phrases, makes Yu-Gi-Oh references, or when you are about to arbitrarily pick between multiple reasonable approaches. Prefer over ask-questions-if-underspecified when the user's tone is casual or playful rather than precision-seeking.
net-ops
IncludedCross-platform network troubleshooting (Windows, macOS, Linux) via local or remote shell. Use for: DNS broken, can't resolve hostnames, nslookup/dig works but apps fail, NRPT, WFP, scutil, /etc/resolver, systemd-resolved, /etc/resolv.conf, NetworkManager, VPN DNS leak residue (ProtonVPN/Mullvad/WireGuard/AnyConnect), AV/firewall blocking DNS or DoH, Tailscale DNS interaction, intermittent connectivity, remote diagnostics over SSH.