release-pipeline-audit
Audit the tag-triggered release pipeline for reproducibility, provenance, signing, supply-chain integrity, and rollback.
Skill by ShaileshRawat1403
Chalk Labs
Need help deploying release-pipeline-audit?
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/shaileshrawat1403-release-pipeline-audit
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
Use cases: data, security
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 25, 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/release-pipeline-audit
What's inside
- Skill name
release-pipeline-audit- Description
- Audit the tag-triggered release pipeline for reproducibility, provenance, signing, supply-chain integrity, and rollback. Returns impact-ordered findings tables.
- License
- MIT
- Instructions
- 72 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
# Release Pipeline Audit Use this skill when the task is to review the path from a release tag to a published artifact. Scope is narrower than [ci-cd-audit](../ci-cd-audit/SKILL.md) — it only covers the workflow(s) that build, sign, and publish releases. ## Use when - the user is about to cut a release and wants a pre-flight check on the pipeline - a release shipped but something went wrong (wrong artifact, missing signature, drifted version) - the project is adding provenance, attestation, or signing for the first time - supply-chain compliance (SLSA, SBOM, signed artifacts) is on the roadmap ## Workflow 1. **Identify the release pipeline.** Find the workflow that runs on tag push (`on: push: tags: ['v*']`) or release publish event. If there are multiple, list them and which artifact each owns. 2. **Trace the tag → artifact chain.** Walk from tag-trigger to upload step. Record: source checkout (with what ref), build step, sign step, upload destination, retention. 3. **Reproducibility.** - Is the build hermetic (locked dependencies, pinned toolchain, no network beyond the package registry)? - Are version strings derived from the tag, or could they drift? - Does the same input produce the same output? Flag wall-clock, hostname, or random-seed inputs in artifact names. 4. **Provenance.** - Is there a SLSA provenance attestation, SBOM, or build receipt artifact? - Is the provenance signed with a key the consumer can verify? - For DAX: look for `release-provenance.json`, `proof-pack/`, and `determinism-proof.json` artifacts written by `bun run release:check`. 5. **Signing.** - Are release archives signed (cosign, gpg, sigstore)? - Is the signing key pulled from secrets correctly (not echoed, not stored on disk after use)? - Is the public key or trust root documented in `SECURITY.md` or a comparable file? 6. **Checksums.** - Does every uploaded artifact have a published checksum (`SHA256SUMS`, per-file `.sha256`)? - Is the checksum file itself signed? 7. **Package-manager publishing.** If the pipeline publishes to npm/cargo/homebrew/winget, check that publishing is gated on signature verification and not on tag presence alone. 8. **Rollback / yank story.** If a bad release ships, can it be yanked? Is there a documented procedure? Are previous releases preserved for downgrade? 9. **Version surface consistency.** The tag, the package manifest (`package.json`, `Cargo.toml`), the `CHANGELOG.md` entry, and any release notes doc must all agree. Flag drift. 10. **Pre-release vs final.** If the pipeline supports prereleases, confirm the gate that distinguishes them (env var, tag suffix) is wired correctly and not bypassable. 11. **Permissions.** The release workflow's `permissions:` block should be minimal — `contents: write` for the release, nothing else unless justified. ## Checks - Do not flag missing SLSA / SBOM unless the project has declared an intent to ship them - Respect project-specific provenance schemes (DAX uses its own determinism-proof artifact) - If publishing is manual (operator runs a command locally), state that explicitly in the verdict — the audit cannot fully cover what happens off-CI - For DAX specifically, the truth surfaces are: the release tag, the top-of-CHANGELOG entry, and `docs/releases/v<version>.md` — all three must agree ## Output contract Findings follow [docs/skills/OUTPUT_CONTRACT.md](../../docs/skills/OUTPUT_CONTRACT.md). Return: 1. **Verdict** — one line + a small table: - `Release workflow(s)`, `Triggers on`, `Artifacts published`, `Signed?`, `Provenance?` 2. **Counts** — Critical/High/Medium/Low/Info totals 3. **Findings** — one table per category, ordered by impact: - Reproducibility - Provenance and attestation - Signing and checksums - Version surface consistency - Rollback story - Workflow permissions 4. **Open questions / assumptions** — manual steps that the workflow does not cover 5. **Residual risk** — anything in the release path that this audit could not verify (e.g., the maintainer's local signing key) 6. **Next actions** — concrete fixes in order ## Evidence to collect - Workflow file path + the tag-trigger line - The exact build command(s) that produce the artifact - The signing step and the key source (env var name, secret name) - The version-string source (tag, package manifest, or computed) - Any provenance/SBOM artifact path written by the pipeline
From ShaileshRawat1403/dax, MIT.
More from ShaileshRawat1403/dax
- 0.619score
mcp-server
ShaileshRawat1403
Create high-quality MCP (Model Context Protocol) servers that extend DAX's tool capabilities.
SkillAFreshness: A★ 5Claude Code +9 - 0.619score
skill-builder
ShaileshRawat1403
Create and improve DAX skills.
SkillAFreshness: A★ 5Claude Code +9 - 0.616score
agentic-loop-audit
ShaileshRawat1403
Audit an agent's main loop for halting conditions, mutation budgets, loop breakers, sensitive-path guards, error recovery, and replayability.
SkillAFreshness: A★ 5Claude Code +9 - 0.616score
artifact-audit
ShaileshRawat1403
Inspect release artifacts, generated outputs, screenshots, and manifests for completeness, consistency, and operator value.
SkillAFreshness: A★ 5Claude Code +9 - 0.616score
ci-cd-audit
ShaileshRawat1403
Audit CI/CD workflows for correctness, caching, action pinning, secret hygiene, and job-graph health.
SkillAFreshness: A★ 5Claude Code +9 - 0.616score
db-migration-review
ShaileshRawat1403
Review a database schema migration for production safety: lock contention, backfill strategy, online vs offline, rollback path, foreign-key timing…
SkillAFreshness: A★ 5Claude Code +9 - 0.616score
deploy-readiness
ShaileshRawat1403
Production readiness audit for a service: health checks, rollback strategy, runbook presence, observability wiring, capacity, dependencies.
SkillAFreshness: A★ 5Claude Code +9 - 0.616score
eval-suite-audit
ShaileshRawat1403
Audit an AI eval suite for coverage, leakage between datasets, deterministic vs stochastic measurement, ladder correctness, and signal quality.
SkillAFreshness: A★ 5Claude Code +9 - 0.616score
gh-actions-security
ShaileshRawat1403
Security-focused review of GitHub Actions workflows: pull_request_target misuse, expression injection, privileged token misuse, supply-chain exposure.
SkillAFreshness: A★ 5Claude Code +9 - 0.616score
observability-audit
ShaileshRawat1403
Audit observability for a service or system: structured logging, metrics for golden signals, distributed tracing, dashboard quality, alert noise, SLO…
SkillAFreshness: A★ 5Claude Code +9 - 0.616score
prompt-audit
ShaileshRawat1403
Audit prompts and system instructions across a repo for hygiene: hardcoded model assumptions, secret leakage, drift between surfaces, length budget, injection…
SkillAFreshness: A★ 5Claude Code +9 - 0.616score
release-readiness
ShaileshRawat1403
Check whether a DAX release is coherent, verifiable, and safe to ship across package, docs, diagnostics, and release artifacts.
SkillAFreshness: A★ 5Claude Code +9
Signals
- Stars
- 5
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- May 16, 2026 (5 months ago)
- Last release
- Oct 8, 2026 (23 days ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/ShaileshRawat1403/dax
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from ShaileshRawat1403
- 0.619score
skill-builder
ShaileshRawat1403
Create and improve DAX skills.
SkillAFreshness: A★ 5Claude Code +9 - 0.619score
mcp-server
ShaileshRawat1403
Create high-quality MCP (Model Context Protocol) servers that extend DAX's tool capabilities.
SkillAFreshness: A★ 5Claude Code +9 - 0.616score
agentic-loop-audit
ShaileshRawat1403
Audit an agent's main loop for halting conditions, mutation budgets, loop breakers, sensitive-path guards, error recovery, and replayability.
SkillAFreshness: A★ 5Claude Code +9 - 0.616score
ci-cd-audit
ShaileshRawat1403
Audit CI/CD workflows for correctness, caching, action pinning, secret hygiene, and job-graph health.
SkillAFreshness: A★ 5Claude Code +9 - 0.616score
db-migration-review
ShaileshRawat1403
Review a database schema migration for production safety: lock contention, backfill strategy, online vs offline, rollback path, foreign-key timing…
SkillAFreshness: A★ 5Claude Code +9 - 0.616score
artifact-audit
ShaileshRawat1403
Inspect release artifacts, generated outputs, screenshots, and manifests for completeness, consistency, and operator value.
SkillAFreshness: A★ 5Claude 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 release-pipeline-audit 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
release-pipeline-audit skill. Agent Mart. https://agentmart.dev/skills/shaileshrawat1403-release-pipeline-audit (last updated 2026-10-24)