agentleFS
ForgeyClapForgeyClap/claude-forge/.claude/skills/ship-readiness/SKILL.md

ship-readiness

Skill2 starsChanged 36 days agoMITView on GitHub

Advisory pre-deploy/pre-handoff checklist (not a blocker). Use when preparing a project for deployment, release, or client handoff — deploy, ship, go live, release, hand off.

  • Reads credentials

3 other projects use a version of this, 1 of them edited.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

5 rules. 0 are common, 4 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 3 edited versions we read keep these.

  • 4

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Before declaring a run complete, run node .claude/forge-bin/forge-verify.cjs <run_id> and report its result.

    line 11
  2. Distinctive:

    No test mode, mock backend, or fake/sample data left in.

    line 12
  3. Distinctive:

    No localhost URLs, debug/test banners, or debug-only UI.

    line 13
  4. Distinctive:

    No unfinished placeholder content/copy.

    line 16
  5. Distinctive:

    No certainty claims; confidence/risk labels on outputs.

    line 43

“Common” is measured: kept by at least 6 in 10 of the 3 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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Ship-readiness (advisory checklist — not a blocker)

This is a practical checklist, not an enforced gate. Report items as pass/fail with evidence — never assert without checking. Nothing here blocks a build; it just helps you not hand off something broken. Production deploys should still have explicit user approval.

Universal

  • Before declaring a run complete, run node .claude/forge-bin/forge-verify.cjs <run_id> and report its result; a mismatch means rework, not a pass.
  • No test mode, mock backend, or fake/sample data left in.
  • No localhost URLs, debug/test banners, or debug-only UI.
  • Secrets in env; .env.example uses placeholders; no real secrets committed.
  • Runtime artifacts swept for leaked secrets too, not just source: node .claude/forge-bin/forge-secret-scrub.cjs [--json] (advisory, reports location only) — covers .claude/forge-runs/*/events.jsonl, which the doctor's git-tracked leak scan cannot see because it is gitignored.
  • No unfinished placeholder content/copy.
  • Forms submit and validate; error handling present on all I/O.
  • Mobile and desktop layout intact; primary CTA clear.
  • Optional: for important code, run the codex-reviewer agent (Codex) and/or security-reviewer — advisory, not required.

Website / landing page

Perf/SEO/a11y pass; meta/OG + canonical; favicon; 404 route; no broken links; responsive at ~360/768/1280; no Lorem/test copy; CTAs/links resolve.

Full-stack app / auth

Auth enforced server-side on protected routes; input validation per endpoint; secrets out of the client bundle; DB migrations reversible; e2e green.

API integration

Retries/backoff + timeout; idempotent writes; consistent error envelope; pagination; webhook signature verification; safe logging.

Dashboard / data viz

Empty/loading/error states; correct aggregation (no silent NULL drops); timezone correctness; no PII leaked client-side.

n8n

Webhook method correct; input schema validated; credentials set (not hardcoded); error branch exists; retry/backoff defined; test/prod flows separated; validate_workflow clean. (Validation is good practice before calling it production-ready.)

Scraping / lead-gen

Public/official-API/permitted sources only; rate limits + robots respected; no login/paywall bypass; data storage lawful + minimal; outreach drafted only (no bulk send without explicit user confirmation).

AI / RAG / chatbot

Clear fallback for empty/low-confidence retrieval; source-aware answers; no hallucinated business facts; logging + human-handoff defined.

Prediction systems

No certainty claims; confidence/risk labels on outputs; data-source quality verified; backtest shown with sample size; no automatic real-money betting.

Telegram bot

Token in env only; webhook vs long-poll chosen; command allowlist; rate-limit; admin-only commands gated; graceful failure message.

Business automation

Dry-run vs live separated; no bulk-send without confirmation; audit log; rollback/dead-letter for failed calls.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.