finish
Complete a feature — merge the approved PR, record plan completion, route the homolog→main decision.
Skill by bravros
Chalk Labs
Need help deploying finish?
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/bravros-finish
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
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 and code. Checked Oct 26, 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/finish~/.bravros/skills
What's inside
- Skill name
finish- Description
- Complete a feature — merge the approved PR, record plan completion, route the homolog→main decision. Use on `/finish` or "finish the feature".
- License
- MIT
- Instructions
- 23 lines in SKILL.md · view on GitHub
Files (4)
- SKILL.md
- briefing.md · references/
- flow.md · references/
- skill.yaml
SKILL.md (first lines)
# finish Read [briefing.md](references/briefing.md) on demand for detailed context and instructions. INTENT: land this feature — merge the PR into its base, record completion in `.planning/events.jsonl`, route the homolog→main decision. Git/project operation only; never touches application code. ## Quick Summary 1. **Resolve PR & Base**: Determine PR and target base (`homolog` or `main`), then drop a **stale review stamp** (Step 1b) — after a multi-round `/address-pr` it still names round 1's commit. 2. **Close Plan**: Record `completed` event in `.planning/events.jsonl`. 3. **CI Check**: `gh pr checks --watch --fail-fast` **redirected to a file**, then `RC=$?` — never piped. Then the readiness gate: merge only at `mergeStateStatus: CLEAN`. 4. **Merge & Verify**: Execute merge gate and post-merge blob verification. 5. **Sync & Clean**: Fast-forward local branches and sweep review stamps. 6. **Main Route**: Route the homolog→main decision with operator confirmation — the main PR repeats step 3 in full. Ask it under a **`Main merge`** header with all three options, and **never use the word "promote" toward the operator here**: this path merges through its own PR gate and needs no token — the police **staging lane** lets a `CLEAN` homolog→main PR merge — but the word sends them off to mint one (afterpay #395/#396 — minted mid-merge, expired unused). `/promote` belongs only in the *defer* option, where it is the real next path. A `✋🏽 Police Block` names its reason: relay it in one line and follow the table in `references/flow.md` Step 7 (UNKNOWN → wait, then retry the same command) — never `gh api`/raw HTTP (B-0036), never `/promote`. 7. **Merge hygiene**: the merge line carries the **literal** PR number (`gh pr merge 1234 --merge`) — the police hook reads raw command text, so a `$VAR` or backtick in the PR argument is unreadable → blocked; no `cd … &&` unless it targets the exact session cwd, never `| tail`; redirect to a file, capture `RC`, inspect. Lock-acquire / merge / `RC` / lock-release may share one Bash call. 8. **Main-merge preflight (before ANY `gh pr merge` into `main`)**: run `bravros police status` (read-only; always exits 0 — read the text: `Police token is MISSING or INVALID` = no token) and `gh pr view 1234 --json mergeStateStatus -q .mergeStateStatus` with the literal number. `CLEAN` + head `homolog` ⇒ the staging lane passes without a token. Not CLEAN (e.g. `UNSTABLE`) ⇒ name the failing check and stop — never attempt the merge. Head is not `homolog` (feature → main) and the token is missing ⇒ ONE ask: "run `bravros police unlock` in a separate terminal, then tell me"; wait — never run `merge-lock acquire … ; gh pr merge` hoping it passes. 9. **Last line, always**: `main @ <sha>` or `homolog only — production pending` — the operator asks "merged to main?" after every run. Refer to [`references/flow.md`](references/flow.md) for full shell script flow details. Its bash is copy-paste code, not illustration: a shell-trap table, the stamp-freshness block, the CI and readiness gates, and the blob verification all have to run **verbatim**.
From bravros/bravros, MIT.
More from bravros/bravros
- 0.609score
after-merge
bravros
Generate an after-merge.md operational deploy checklist for homolog-to-main aggregate releases.
SkillAFreshness: A★ 3Claude Code +9 - 0.609score
laravel-pest5-upgrade
bravros
Upgrade a PHP/Laravel repo from Pest 4 to Pest 5 and turn on Test Impact Analysis (TIA) so local full runs only re-execute the tests a change can reach.
SkillAFreshness: A★ 3Claude Code +9 - 0.609score
interview-me
bravros
Stress-test a plan or design via a round-by-round interview — ask the whole frontier of independent questions each round until every branch of the decision…
SkillAFreshness: A★ 3Claude Code +9 - 0.609score
doctor-plus
bravros
Runs Claude Code's built-in /doctor health check, then audits the workspace against Anthropic's 6 then-and-now context-engineering shifts (rules→judgement…
SkillAFreshness: A★ 3Claude Code +9 - 0.609score
context
bravros
Scan project and generate/audit CLAUDE.md files with stack auto-detection and Context7 docs.
SkillAFreshness: A★ 3Claude Code +9 - 0.606score
advise-project-approach
bravros
Research and advise how a software project should be built — stack selection, architecture, operating-cost tradeoffs, and real-world comparables.
SkillAFreshness: A★ 3Claude Code +9 - 0.606score
graphify-this-project
bravros
Set up a graphify knowledge graph for the current project — in-project committed graph.json, AST extraction + external labelling, tracked refresh hooks…
SkillAFreshness: A★ 3Claude Code +9 - 0.606score
backlog
bravros
Capture, list, and promote pre-planning ideas.
SkillAFreshness: A★ 3Claude Code +9 - 0.606score
batch-merge-prs
bravros
Verify N PRs, address review feedback, ordered-merge to the staging branch (never main), close linked issues/backlog per merge, then hand off the full test…
SkillAFreshness: A★ 3Claude Code +9 - 0.606score
graphify-status
bravros
Report knowledge-graph label coverage across every graphify-enabled project on this machine — communities, labels, and how many nodes still render as…
SkillAFreshness: A★ 3Claude Code +9 - 0.606score
laravel-pest-rector
bravros
Adopt Pest's Rector plugin on a Pest 5 project and refactor tests into Pest matchers — scoped to tests/, dry-run gated.
SkillAFreshness: A★ 3Claude Code +9 - 0.603score
hotfix
bravros
Emergency hotfix deploy — commit, push homolog, PR to main, merge now.
SkillAFreshness: A★ 3Claude Code +9
Signals
- Stars
- 3
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Oct 1, 2026 (29 days ago)
- Last release
- Oct 1, 2026 (29 days ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/bravros/bravros
- Website
- bravros.dev
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from bravros
- 0.609score
doctor-plus
bravros
Runs Claude Code's built-in /doctor health check, then audits the workspace against Anthropic's 6 then-and-now context-engineering shifts (rules→judgement…
SkillAFreshness: A★ 3Claude Code +9 - 0.609score
laravel-pest5-upgrade
bravros
Upgrade a PHP/Laravel repo from Pest 4 to Pest 5 and turn on Test Impact Analysis (TIA) so local full runs only re-execute the tests a change can reach.
SkillAFreshness: A★ 3Claude Code +9 - 0.609score
context
bravros
Scan project and generate/audit CLAUDE.md files with stack auto-detection and Context7 docs.
SkillAFreshness: A★ 3Claude Code +9 - 0.609score
after-merge
bravros
Generate an after-merge.md operational deploy checklist for homolog-to-main aggregate releases.
SkillAFreshness: A★ 3Claude Code +9 - 0.609score
interview-me
bravros
Stress-test a plan or design via a round-by-round interview — ask the whole frontier of independent questions each round until every branch of the decision…
SkillAFreshness: A★ 3Claude Code +9 - 0.606score
advise-project-approach
bravros
Research and advise how a software project should be built — stack selection, architecture, operating-cost tradeoffs, and real-world comparables.
SkillAFreshness: A★ 3Claude 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 finish 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
finish skill. Agent Mart. https://agentmart.dev/skills/bravros-finish (last updated 2026-10-24)