agentleFS
zen-mcpzen-mcp/browser/.agents/skills/capture-evidence-bundle/SKILL.md

capture-evidence-bundle

Skill2 starsChanged 7 months agoApache-2.0View on GitHub

Capture browser task evidence by collecting page snapshot text, screenshots, current URL/title, and artifact metadata. Use when a task needs auditable proof of the page state before or after browser actions.

The file

We split this skill into the rules it gives an agent

8 rules. About 0 are common, 2 are distinctive, and 6 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 6

    Very specific

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

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

    Use this skill when a user asks for proof, screenshots, page evidence, a before and after capture, or a reproducible record of a browser …

    line 15
  2. Very specific:

    Call get_page_snapshot to capture URL, title, and visible text.

    line 26
  3. Very specific:

    Call take_screenshot with a sanitized label.

    line 27
  4. Very specific:

    If a specific UI element matters, capture a second screenshot with selector.

    line 29
  5. Very specific:

    Call list_artifacts and record the new screenshot file names.

    line 31
  6. Distinctive:

    Never include passwords, auth tokens, cookies, or raw headers in summaries.

    line 37
  7. Distinctive:

    Never persist credentials in artifacts.

    line 38
  8. Very specific:

    Never bypass take_screenshot or file tools with ad hoc filesystem paths.

    line 39

≈ “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

Capture Evidence Bundle

When To Use

Use this skill when a user asks for proof, screenshots, page evidence, a before and after capture, or a reproducible record of a browser task.

Inputs

  • label: short evidence label used as a screenshot prefix.
  • full_page: whether the screenshot should include the full page.
  • selector: optional element selector for focused screenshots.

Procedure

  1. Call get_page_snapshot to capture URL, title, and visible text.
  2. Call take_screenshot with a sanitized label. Use full_page=True for reports, audits, and long pages.
  3. If a specific UI element matters, capture a second screenshot with selector.
  4. Call list_artifacts and record the new screenshot file names.
  5. Return a concise summary with URL, title, screenshot names, and any notable page text needed to explain the result.

Never

  • Never include passwords, auth tokens, cookies, or raw headers in summaries.
  • Never persist credentials in artifacts.
  • Never bypass take_screenshot or file tools with ad hoc filesystem paths.

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.