agentleFS
jqaisystemsjqaisystems/jqai-ai-skills/skills/website-change-verifier/SKILL.md

website-change-verifier

Skill3 starsChanged 6 months agoMITView on GitHub

Verify website changes across source, routes, rendering, metadata, discovery surfaces, and deployment state. Use when a user asks to check a new or edited page, confirm an exact replacement, validate a route, inspect responsive behavior, verify metadata or sitemap changes, compare localhost with production, or determine whether a website change is fully working.

The file

We split this skill into the rules it gives an agent

12 rules. About 0 are common, 10 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈10

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Verify the complete user-visible result of a website change.

    line 8
  2. Distinctive:

    Read references/framework-checks.md for the detected stack.

    line 14
  3. Distinctive:

    Check which parts of this chain apply:

    line 19
  4. Distinctive:

    Read references/verification-matrix.md and select checks proportional to the change.

    line 33
  5. Distinctive:

    Run syntax, type, lint, test, or production-build checks supported by the project.

    line 34
  6. Very specific:

    Use scripts/verify_routes.py for deterministic status and response-text checks.

    line 36
  7. Distinctive:

    Check discovery surfaces only when the content is intended to appear there.

    line 39
  8. Very specific:

    Use --expect-status "URL::404" for an intentionally missing route and --json for machine-readable output.

    line 50
  9. Distinctive:

    Do not include authentication values in URLs, commands, logs, or reports.

    line 55
  10. Distinctive:

    Do not deploy, publish, submit indexing requests, change DNS, purge caches, or modify external services unless explicitly authorized.

    line 56
  11. Distinctive:

    After an authorized deployment, repeat the relevant checks against the final public URL.

    line 57
  12. Distinctive:

    Use one of these states:

    line 61

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 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

Website Change Verifier

Verify the complete user-visible result of a website change. Default to read-only checks and report the highest completion state supported by evidence.

1. Identify the change and platform

  • Confirm the project root, target route, requested outcome, source of truth, and expected visibility.
  • Inspect project instructions, manifests, entrypoints, routing, and nearby implementation patterns.
  • Read references/framework-checks.md for the detected stack.
  • Preserve existing public URLs and unrelated behavior unless the request explicitly changes them.

2. Map affected surfaces

Check which parts of this chain apply:

  1. content or data source
  2. route or page rendering
  3. navigation, listings, related content, and internal links
  4. title, description, canonical, Open Graph, structured data, and accessibility text
  5. sitemap, robots rules, llms.txt, feeds, search indexes, or static export
  6. images, video, downloads, and responsive variants
  7. local server, deployment, cache, and live state

For hidden content, distinguish navigation removal from direct access, sitemap presence, and search visibility.

3. Verify in layers

  • Read references/verification-matrix.md and select checks proportional to the change.
  • Run syntax, type, lint, test, or production-build checks supported by the project.
  • Confirm the correct listener and make a real request to each affected route.
  • Use scripts/verify_routes.py for deterministic status and response-text checks.
  • Inspect rendered desktop and mobile behavior when layout, interaction, forms, or media changed.
  • Confirm required text is present and replaced text is absent.
  • Check discovery surfaces only when the content is intended to appear there.

Example:

python scripts/verify_routes.py `
  --url http://127.0.0.1:8766/example `
  --contains "http://127.0.0.1:8766/example::Expected heading" `
  --not-contains "http://127.0.0.1:8766/example::Old wording"

Use --expect-status "URL::404" for an intentionally missing route and --json for machine-readable output.

4. Respect authority boundaries

  • Access only the project and URLs placed in scope.
  • Do not include authentication values in URLs, commands, logs, or reports.
  • Do not deploy, publish, submit indexing requests, change DNS, purge caches, or modify external services unless explicitly authorized.
  • After an authorized deployment, repeat the relevant checks against the final public URL.

5. Report the supported completion state

Use one of these states:

  • implemented: source changes exist but user-visible verification is incomplete
  • verified locally: build and local route or browser checks pass
  • deployed: an authorized deployment completed but live verification remains
  • verified live: the production URL and intended behavior were checked after deployment

Return the checked routes, expected and observed results, build or syntax results, browser coverage, discovery-surface status, highest completion state, and exact remaining blocker.

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.