agentleFS
suryastsuryast/free-ai-agent-skills/static-site-release-verification/SKILL.md

static-site-release-verification

Skill2 starsChanged 35 days agoMITView on GitHub

Verifies source, build artifacts, deployed revision, and clean-URL asset parity. Use when releasing a static site, checking CDN freshness, or investigating a successful upload with stale production content.

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Verify source-to-build content and links before deployment.

    line 14

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “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

Static-site release verification

Establish the release contract

  1. Identify the repository, source revision, actual build command, output directory, deployment target, and authorized scope. Read the host's deployment provenance without changing it. Keep expected source revision and observed deployed revision separate; an upload receipt or branch tip is not deployment evidence.
  2. Build the reviewed revision in an isolated directory. Record dirty/untracked inputs and build tool versions. Select changed routes plus representative template routes (all routes for site-wide changes). Record exact expected content needles, canonical URLs, and referenced CSS/JS/data files from that artifact.
  3. Verify source-to-build content and links before deployment. Do not claim a committed artifact was tested if inputs changed after the build. Preserve raw artifacts; normalize only explicitly documented host injections, never arbitrary differences.
  4. If deployment is separately approved, execute it through the documented interface. This skill's checker never deploys or purges. Read back the host's exact deployed revision or immutable release ID and tie it to the tested artifact. If provenance is absent, mark revision verification unavailable; do not infer it from matching text.
  5. Check immutable preview and ordinary canonical production URLs separately. Use clean URLs without cache-busting queries. Compare HTML content/canonical and every selected CSS/JS/data file against local build bytes. Fresh HTML can reference stale unversioned assets. Record status, redirects, hashes, cache headers, timestamp, and tested vantage point.
  6. A stale edge is a release hold, not permission to purge. Request a scoped purge only if authorized; recheck the same clean URLs afterward. A passing single-edge probe is not proof of global CDN propagation. Keep content success distinct from provenance success and optional visual/accessibility checks.

Optional read-only checker

Use check_release.py with a reviewed manifest modelled on example.json. Paths are relative to --build-dir; routes and asset URLs are same-origin clean paths. List assets explicitly, including data and nonstandard HTML references: the checker does not crawl the whole site or discover dependencies. HTML verification uses UTF-8 needles and exactly one canonical; assets require exact byte equality. HTML byte parity is not required because hosts may inject markup.

python3 static-site-release-verification/scripts/check_release.py \
  --manifest release.json --build-dir dist --base-url https://example.com \
  --observed-revision "$HOST_REPORTED_REVISION" > release-evidence.json

Obtain HOST_REPORTED_REVISION from independent hosting provenance, not by copying the expected value. The sample's revision is synthetic. A missing/mismatched revision or failed check exits nonzero. The checker rejects redirects rather than silently verifying another host, bounds response size and request count, and sends no authorization headers. It may expose manifest routes to the named server; inspect scope before running. It does not execute downloaded code.

Handoff

Report expected/observed revision and provenance source, build evidence, tested URLs/assets and vantage point, failed assertions, remaining route/edge coverage, and exact next approved operation. Keep uploads, deployments, and verified releases distinct. Never turn an approval timeout into alternate-tool deployment or repeated purge attempts.

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.