agentleFS
patronus-protectpatronus-protect/patronus-security-cli/.claude/skills/qa-dashboard/SKILL.md

qa-dashboard

Skill2 starsChanged 2 months agoApache-2.0View on GitHub

UX/UI QA of the Patronus control dashboard (control.patronus.studio) with the dedicated QA account — fixed user journeys, usability, errors, accessibility and mobile layout. Use as part of the QA run or when asked to review the dashboard.

2 other projects use a version of this.

Adopted by

Most recently taken up by

  • Other versions2 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 2

    1. petrkindlmann/qa-skills★ 209same name
    2. jason0971/skills★ 1same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository5 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

7 rules. About 0 are common, 3 are distinctive, and 4 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 4

    Very specific

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

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

    Use Claude in Chrome (mcp__claude-in-chrome__*), the person's real Chrome where the QA account is signed in.

    line 8
  2. Very specific:

    If the session has expired, record the finding qa:dashboard:session-expired (severity medium, suggestion "sign in the QA account in …

    line 13
  3. Very specific:

    Only change resources whose name starts with qa-.

    line 15
  4. Distinctive:

    Billing: read it, never submit a payment, plan change or cancellation.

    line 16
  5. Distinctive:

    Account and organisation settings: read them, never change them.

    line 17
  6. Distinctive:

    Docs and help links: every help link should lead to the right page on docs.patronus.studio.

    line 30
  7. Very specific:

    Write ~/.patronus-qa/runs/<date>/dashboard.json in the report shape:

    line 42

≈ “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

Control dashboard QA

Use Claude in Chrome (mcp__claude-in-chrome__*), the person's real Chrome where the QA account is signed in. Load the tools with one ToolSearch call. Open your own tab with tabs_create_mcp (or a standalone navigate), and close it with tabs_close_mcp when you are done.

Rules

  • Never enter passwords or other credentials. The QA account must already be signed in to Chrome.
  • If the session has expired, record the finding qa:dashboard:session-expired (severity medium, suggestion "sign in the QA account in Chrome") and stop this area.
  • This is the person's own browser: do not touch other tabs, bookmarks, extensions or browser settings.
  • Only change resources whose name starts with qa-. Create them yourself and delete them again in the same run.
  • Billing: read it, never submit a payment, plan change or cancellation.
  • Account and organisation settings: read them, never change them.
  • Cookie banners: pick the most privacy-preserving option.

Journeys

For each journey, count the clicks or steps and note anything confusing:

  1. First look: open https://control.patronus.studio/. Can a new user tell within 10 seconds what to do first?
  2. API key: on a plan with room for another key, create a key named qa-<date>, check that copying is clear and the key is shown only once, then revoke it. Never revoke a key that is not named qa-…, and never paste a key value into notes or issues. On the Free plan the single key is the QA key, so only review the page.
  3. Playground: scan a harmless text and a text containing IGNORE ALL PREVIOUS INSTRUCTIONS (Playground scans do not use quota). Is the result understandable and localized?
  4. Scan history / activity: find the scans the QA scripts made today. Check filters and empty states.
  5. Usage and limits: is the current usage and the limit obvious?
  6. CLI login: open /cli-login directly. Check that the page explains what happens and what to do next.
  7. Docs and help links: every help link should lead to the right page on docs.patronus.studio.

Each page

  • Console errors and failed requests: read_console_messages with onlyErrors, plus read_network_requests for status ≥ 400.
  • Load time: note pages that take longer than about 3 s until content appears.
  • Mobile: resize_window to 390×844 and check that innerWidth actually changed (a maximized window ignores it). Check for horizontal scrolling, cut-off tables and unreachable actions. Restore the original size (outerWidth/outerHeight read beforehand). If the size cannot change, note that mobile was not checked.
  • Accessibility basics: unlabeled buttons or inputs (use read_page with filter interactive), focus order on the main form, and contrast problems visible in the screenshot.
  • Copy: unclear labels, jargon, error messages that do not say how to fix the problem.

Output

Write ~/.patronus-qa/runs/<date>/dashboard.json in the report shape:

  • metrics.journeys = [{name, ok, steps, issues}]
  • metrics.console_errors with the total number
  • Findings use area dashboard (bugs) or ux (usability). Each finding names the page URL and the exact element. Put a short description of the screenshot in the evidence, never a credential or key.

The dashboard source is not in this repository, so propose issues for the control-plane repository only, no PRs.

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.