SKILLEMALL.ai
GitHub ↗ RU Sign in

BB vector

Automating browsers via Playwright and Chrome DevTools for data collection, form interaction, screenshot capture, and network monitoring. Task completion focus (vs Voyager for E2E testing).

simota/agent-skills Agent Skills author: simota 11 files body ≈ 5 886 tokens Open the sourcegithub.com↗ analyzed 3 h ago

Automating browsers via Playwright and Chrome DevTools for data collection, form interaction, screenshot capture, and network monitoring.

As a process B 66/100 · Nearly there — weak spots: inputs and preconditions, running it twice

ProcedurePlaywrightInfrastructureAI and agentsMedia and videotype and topics are labelled automatically from the skill text
Runs in: Claude / Claude Code ChatGPT / Codex Gemini CLI Cursor GitHub Copilot / VS Code Qwen Code Kimi Code CLI Z.ai (GLM) DeepSeek (Deep Code) Mistral Vibe LangChain (deepagents) OpenClaw Goose OpenCode
JSON
Technical rating
B
81/100
safety, quality, tests
Safety 60%
92
Quality 40%
65
Run on models
none yet
Process rating
B
66/100
Nearly there
Inputs and preconditions w 11
0
Running it twice w 4
30
When it triggers w 12
50
the three weakest of ten parameters · all ten

What is at stake

Medium-severity findings: the skill is probably honest, but read what alarmed the scanner.

Exfiltration medium severity

Below is the worst case for this category. The finding here is medium: the guard saw a sign, not a proof.

If you install

The instructions or scripts send data out: environment variables, keys, file contents, chat history. You may never notice, because the agent performs the upload as "part of the task".

For the author

If the upload is not needed for the task, remove it: catalogs flag such skills and delist them. If it is needed, name the destination explicitly, say what leaves the machine, and give the user a switch.

How to improve

  1. Say in the description WHEN to use the skill ("use when…", example requests): that is the agent's main cue.
  2. The SKILL.md body is over 5,000 tokens: move reference detail into references/ and load it when needed.
For the model run — optional
  • Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
  • A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.

Guard findings · 4

✓ No critical or high findings

Medium and low: 4
  • medium Exfiltration intent-browser-credential-store reference/parallel-sessions.md:19
    Accesses a browser credential / cookie store (detector / deny-list definition)
    **Anti-pattern**: opening N pages in one context for "parallelism" — they share cookies/cache/IndexedDB and serialize on the same JS event loop for blocking operations. **One context per worker** is t
    detector
  • low Risky intent intent-offensive-security reference/stealth-mode.md:15
    Offensive-security / dual-use content (legitimate for authorised testing; review intended use)
    | Use case justification | Research / accessibility / monitoring / authorized pentesting |
  • low Risky intent intent-offensive-security reference/stealth-mode.md:243
    Offensive-security / dual-use content (legitimate for authorised testing; review intended use)
    Authorization: ___________ (research / accessibility / monitoring / pentesting / OTHER → REFUSE)
  • low Risky intent intent-offensive-security SKILL.md:75
    Offensive-security / dual-use content (legitimate for authorised testing; review intended use)
    - security penetration testing: `Probe`

Files scanned: 11. Evidence is masked. Grey chips explain why severity was lowered.

Against the Agent Skills spec

  • warning description-no-when description does not say WHEN to use the skill (no "use when")
  • warning body-long SKILL.md body ≈ 5886 tokens (recommended < 5000); move details to references/
  • note edit-residue the text marks something as outdated (lines 142): check that old rules are not kept next to new ones — the full check reads the text for contradictions

Process rating: all ten parameters 66/100

  • 0Inputs and preconditions. Does not say what the process needs to start
  • 30Running it twice. 8 mutating operations with no state check
  • 50When it triggers. No condition that starts the skill
  • 50Failures and branches. 0 branches, has a failure section
  • 60Tools and files. Uses tools (bash, web, node) that frontmatter does not declare
  • 70Execution cost. Instruction body is 5886 tokens
  • 100Steps. 126 steps
  • 100Result and completion. Output format and completion criterion are stated
  • 100Consistency. Name and required fields are in place
  • 100Progress reporting. Reports progress
  • medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
  • low 15 top-level sections: this looks like several domains in one skill

Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.

Quality signals

  • +5Description has no quoted example phrases that should trigger the skill
  • +4Description does not say when NOT to use the skill (false activations)
  • +4No input/output examples
  • +1No license
  • +2Single-language instructions
  • +3Description length 189: enough signal without eating the budget
  • +4Structure: 21 headings
  • +3Step-by-step instructions: 126 items
  • +3Output format is stated explicitly
  • +4Reference files are cited in the instructions (9 of 9)

Quality base 70; lint remarks subtract, signals add up to 100. Result: 65.