release-gate-chair
Runs the three pre-launch passes in order — web-preflight for external exposure, pre-ship-security-red-team for application behaviour in an isolated environment, pre-ship-performance-audit for measured speed — records each lane's own verdict and findings into one ledger, and computes a single release posture (HOLD or ready) with a script rather than judgement. Use when someone asks "is this safe to launch", "ready to ship", "should we deploy", "give me a go/no-go", or before any deploy of a site or client app you own or are authorized to test. Three verdicts stay separate; a pass in one lane never covers another. Variant: this fork runs the lanes itself and copies their finding tables verbatim; it is the default release gate. Not for a single-lane check (invoke that skill directly), for reports that already exist (use release-gate-validator), for penetration testing of production, or for performing the deploy.
- License 许可
- MIT
- 最近更新
- 2026-11-14
- Stars
- ★ 0
- Forks
- 0
01 前言 Front Matter
- name
- release-gate-chair
- version
- 0.1.0
- description
- Runs the three pre-launch passes in order — web-preflight for external exposure, pre-ship-security-red-team for application behaviour in an isolated environment, pre-ship-performance-audit for measured speed — records each lane's own verdict and findings into one ledger, and computes a single release posture (HOLD or ready) with a script rather than judgement. Use when someone asks "is this safe to launch", "ready to ship", "should we deploy", "give me a go/no-go", or before any deploy of a site or client app you own or are authorized to test. Three verdicts stay separate; a pass in one lane never covers another. Variant: this fork runs the lanes itself and copies their finding tables verbatim; it is the default release gate. Not for a single-lane check (invoke that skill directly), for reports that already exist (use release-gate-validator), for penetration testing of production, or for performing the deploy.
- argument-hint
- <production URL> [--app <isolated/local URL>] [--project <path>]
02 资源文件 3
- 其他.gitignore28 B
- 参考文档references/ledger.md4.1 KB
- 脚本scripts/check_gate.mjs8.5 KB
03 SKILL.md 正文
Release gate chair
Three lanes, three verdicts, one posture. The gate adds nothing to what the three skills can
detect; what it adds is that the arithmetic is redone by scripts/check_gate.mjs, coverage is
stated, and a lane that did not run cannot be mistaken for a lane that passed.
<skill-dir> is this skill's directory.
Inputs
| Input | Needed for | If absent |
|---|---|---|
| Production URL | exposure lane | Ask. A gate with no target is a template. |
| Isolated or local app URL | app-tests lane | Record the lane untested, or descoped_by with the user's reason. Never run red-team against production. |
| Project path | performance lane | Perf lane runs measurement-only and says so |
Multiple hosts are gated independently and keep separate rows, then any redirect or canonical-host relationship is named.
1. Profile the target, then say so
Classify before dispatching: static-site (no auth, no datastore), web-app (auth, roles, quota,
datastore), api-or-worker (no page to measure in a browser), preview-only (not yet deployed).
Write the profile into the run directory. The profile decides which lanes run in full and which
rows are not-applicable — a static marketing site has no login, role or quota to attack, and
dispatching seven subagents at it is why gates get skipped.
not-applicable is the gate's claim about the target, so the gate writes it and names the profile
that justified it. A lane is never quietly shortened.
2. Run the lanes in order
Do not start the next lane on the strength of the previous one's pass.
Exposure. web-preflight. Bounded GET/HEAD, so a live host is in scope. Emit its host verdict
in its own precedence BLOCK > INCONCLUSIVE > PASS WITH FIXES > PASS, and its PREFLIGHT-<nn>
findings.
App-tests. pre-ship-security-red-team. Local or isolated only, synthetic accounts. Accept
this lane only on mechanical evidence:
node "<red-team-skill-dir>/scripts/check_results.mjs" <run-dir>
node "<red-team-skill-dir>/scripts/check_results.mjs" <run-dir> --require-fixes
Both must exit 0. Record the second one's exit code in the ledger. Findings are REDTEAM-S<nn>.
Perf. pre-ship-performance-audit. Four read-only subagent audits, findings PERF-<nn>,
fixed-verified only with an after-median taken under the same conditions.
3. Record the ledger
.release-gate/<run-id>/ in the audited repo, gitignored first. Format and every column rule in
references/ledger.md.
This fork is the chair: it populates the ledger itself, copying each lane's finding table verbatim — ids, routes, severities, remediation states as that lane wrote them. Never re-summarise a finding, never soften a remediation, never derive one lane's verdict from another's. After copying a lane, count the rows you wrote against the rows in that lane's own report; a mismatch is fixed by re-reading the report, not by adjusting the count.
Fill touches per finding from what the finding says, not from what you hope. Blanket none is a
claim; unknown on a CRITICAL or HIGH row holds the gate, which is the cheaper mistake.
4. Compute the posture
node "<skill-dir>/scripts/check_gate.mjs" .release-gate/<run-id> --sha <current-head-sha>
Exit 0 ready, 1 HOLD, 2 contract break. Report the rule codes that fired — the rule used is
part of the answer, not a footnote. ready means ready for the coverage stated, and that line
is part of the verdict.
A HOLD is not overridden by urgency. Deploying anyway is the user's call, made in the open, and
the gate records that it was made.
5. Deliver, and re-check after deploy
Print the checker's block, then write the same register to docs/release/<YYYY-MM-DD>-<target>-gate.md
and commit it if the user wants it in history. Re-runs reuse finding-ids; an id that changed
meaning is a contract break — say so rather than quietly renumbering.
Then offer the post-deploy re-check: one exposure lane row for the live host in a new run
directory, reusing existing PREFLIGHT-* ids. It can move ready to HOLD. It cannot move a
HOLD to ready, because it adds no coverage the pre-deploy pass lacked.
Does not
- Deploy anything, or widen any lane's scope. New checks belong to the lane's own skill; the gate cites it.
- Merge scores, average severities, or produce one number for three different kinds of claim.
- Turn an unrun lane into a pass, or report a partial run as complete.
- Run red-team against production, or reproduce a secret in the register.
If a caffeinity-stack skill is installed, read it before reporting Caffeinity work (deploy runbook, no-fabricated-evidence).
Brew 3 · Espresso in caffeinity-routing is the sequence this skill implements.
Verify the gate before trusting it
node "<skill-dir>/tests/posture.mjs"
20 fixtures, 7 of which must come back ready. A checker that always answered HOLD would pass the
HOLD cases by accident, so the ready cases are the real test. Run this after any edit to
check_gate.mjs, and before publishing.
Known weakness of this fork
Transcription is model-mediated. The mitigation is verbatim copying plus a row-count reconciliation, which limits the damage but does not eliminate it. If a lane's report and the ledger ever disagree, the report wins and the ledger is rewritten.
04 同仓库更多 Skills
tune-and-ship
Takes one published skill at a time from as-is to tuned, fact-checked, released and live, then moves to the next one in the catalog. Each run tunes the skill with tune-skill, checks every version number, flag, price and command it states against the primary source, re-reads its siblings' real SKILL.md files, polishes the listing with findable-polish, writes the version and release notes, opens one PR, and after the merge confirms the site page and marketplace copy are live. Use whenever the user wants to fine-tune their other skills, work through the catalog, give a published skill the full treatment, fact-check a skill's claims, or ship a tune instead of drafting it, even if they never say "ship". Not for a tune with no release (use tune-skill), drafts with nothing pushed (use tune-and-polish), a new skill (use skill-launch), regressions from the nightly health check (use caffeinity-routing's Descale brew), or a whole collection (use skill-estate).
skill-estate
Takes a whole collection of agent skill repos from scattered to published, consistent and self-maintaining, in gated phases: inventory every repo, reconcile site, marketplace and README listings, test and tune the family, close workflow gaps, wire publishing so new skill repos go live on their own, bundle a plugin with agents, add trigger tests and directory listings, and leave a maintenance loop and a map. Each phase hands off to the skill that owns it and ends in PRs. Use when someone's skills site and GitHub disagree, wants all their skill repos audited or cleaned up, wants new skills to publish themselves, or wants that cleanup repeated on another org, even if they never say estate. Not for one skill (use tune-skill), one listing (use findable), the site pipeline alone (use publish-pipeline), or weekly upkeep (use caffeinity-routing's Descale brew).
paper-trail
Drafts the policy pages a small product needs before it takes money or data (privacy policy, terms of service, refund policy, cookie notice) from a fact sheet about what the product actually collects, stores, shares and charges, then checks the drafts for placeholders, missing sections and facts nobody supplied. Use when someone needs a privacy policy, terms, refund or cookie page for a site, app, SaaS, newsletter or digital product, asks what legal pages they need before launch, or wants existing policy pages checked against what the product really does, even if they never say "legal". Not legal advice and not for contracts, NDAs or employment documents (use a lawyer), and not for the launch plan itself (use solo-income-blueprint) or the site build (use site-prompt).
course-builder
Designs a course for a group of learners (a class, cohort, workshop series, bootcamp or internal training): measurable outcomes, a module sequence with prerequisites, activities and assessments aligned to each outcome, a time budget that fits the hours available, and a rubric, then checks the plan with a script so no outcome goes untaught or unassessed. Use when someone is teaching others and asks for a curriculum, syllabus, course outline, module plan, cohort or workshop structure, or wants an existing course checked for gaps and overload, even if they never say "curriculum". Not for one person's own learning plan (use study-system), for being taught the topic (use topic-primer), or for marketing the course (use sales-copy).
wordsmith
Write, draft, rewrite, edit, proofread, polish and improve text while preserving the author's intended meaning, facts and voice. Use when the user wants clearer, shorter or better-toned writing; grammar, structure or flow fixes; emails, website or marketing copy, headlines, outlines or summaries; a different reading level; an accessibility, inclusivity or accuracy review; or content prepared for publication, even if they never say "edit". Also handles /wordsmith commands such as draft, critique, audit, tighten, tone and final. Not for resumes, CVs or LinkedIn profiles (use recruiter-lens if installed), not for code or code comments, and not for building .docx or .pdf files (write the text here, then use the file skill).
web-preflight
Use when shipping, deploying, publishing, launching, reviewing, or pre-flighting a website or API and you want a lightweight check for accidental exposure of secrets, source or configuration files, admin and debug surfaces, unsafe redirects, missing security controls, and misleading 200 responses. Trigger phrases include "before we ship", "security check", "preflight", "audit the site", "check the production URL", "is my .env exposed", and "why does /.env return 200", including when access-log or analytics evidence is supplied. Checks only sites the user owns or is authorized to assess, with bounded GET and HEAD requests and a fixed compact path set; distinguishes genuine resources from SPA catch-all and blanket WAF responses. Not for penetration testing, exploitation, credential testing, broad port scanning, or application-code review; use a dedicated authorized security-testing or code-review workflow instead.