SKILLEMALL.ai
GitHub ↗ RU Sign in

FF headful-browser-vnc

Headful Chromium with VNC/noVNC operator UI and Chrome CDP exports (cookies, screenshots, outerHTML).

Not recommendedcritical or high security findings · low grade F
ClawHub Agent Skills author: Wenlue Song v1.0.2 MIT-0 15 files · 8 scripts body ≈ 3 563 tokens Open the sourceclawhub.ai↗ analyzed 3 d ago

As a process F 45/100 · Will not run — References files that are not bundled: templates/x11vnc.service.j2, templates/novnc.service.j2

ProcedureInfrastructuretype and topics are labelled automatically from the skill text
Runs in: OpenClaw 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) Goose OpenCode
JSON
Technical rating
F
30/100
safety, quality, tests
Safety 60%
16
Quality 40%
52
Run on models
none yet
Process rating
F
45/100
Will not run
References files that are not bundled: templates/x11vnc.service.j2, templates/novnc.service.j2
Tools and files w 18
0
Result and completion w 14
0
Inputs and preconditions w 11
0
the three weakest of ten parameters · all ten

What is at stake

The skill contains fragments that, in the wrong hands, cost money or data. Below: what the installer risks and what the author should do.

Exfiltration
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. Address the high-severity findings: each costs 18 safety points. If one is a false positive, add the rule id to guard.allow in spec.yaml.
  2. Say in the description WHEN to use the skill ("use when…", example requests): that is the agent's main cue.
  3. The text references files that are not there: add them or drop the references.
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 · 8

  • high Exfiltration intent-browser-credential-store README.md:28
    Accesses a browser credential / cookie store
    - **Programmatic capture**: export rendered `outerHTML`, full-page screenshots, and cookies using Chrome CDP (Playwright/Puppeteer compatible). Exports are intended for downstream automated analysis o
  • high Exfiltration intent-browser-credential-store README.md:90
    Accesses a browser credential / cookie store
    - **Purpose**: export cookies from a running Chrome instance via the Chrome DevTools Protocol.
  • high Exfiltration intent-browser-credential-store SKILL.md:28
    Accesses a browser credential / cookie store
    - **Programmatic capture**: export rendered `outerHTML`, full-page screenshots, and cookies using Chrome CDP (Playwright/Puppeteer compatible). Exports are intended for downstream automated analysis o
  • high Exfiltration intent-browser-credential-store SKILL.md:90
    Accesses a browser credential / cookie store
    - **Purpose**: export cookies from a running Chrome instance via the Chrome DevTools Protocol.
Medium and low: 4
  • medium Exfiltration intent-browser-credential-store README.md:97
    Accesses a browser credential / cookie store (quoted — discussed, not commanded)
    - **Note**: these export helpers (`export_page.sh`, `export_cookies.sh`) are designed to be idempotent and safe to run after manual operator interventions. They place artifacts into `out/<session_id>/
    quoted
  • medium Exfiltration intent-browser-credential-store SKILL.md:97
    Accesses a browser credential / cookie store (quoted — discussed, not commanded)
    - **Note**: these export helpers (`export_page.sh`, `export_cookies.sh`) are designed to be idempotent and safe to run after manual operator interventions. They place artifacts into `out/<session_id>/
    quoted
  • low Dangerous commands cmd-background-process scripts/start_chrome_debug.sh:28
    Starts a background / autostarted process
    setsid nohup /usr/bin/google-chrome --user-data-dir="$PROFILE_DIR" "$PROXY_FLAG" --remote-debugging-port=""$REMOTE_PORT"" --no-first-run --no-default-browser-check --disable-extensions --disable-gpu -
  • low Dangerous commands cmd-background-process scripts/start_vnc.sh:42
    Starts a background / autostarted process
    setsid x11vnc -display "$DISPLAY_VAL" -rfbauth "$VNC_PASS_FILE" -forever -shared -o ""$LOG"" &

Files scanned: 14. 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 missing-ref reference to a missing file: templates/x11vnc.service.j2
  • warning missing-ref reference to a missing file: templates/novnc.service.j2
  • note frontmatter-key unknown frontmatter key "required_env"

Process rating: all ten parameters 45/100

Will not run. References files that are not bundled: templates/x11vnc.service.j2, templates/novnc.service.j2
  • 0Tools and files. 2 referenced file(s) missing: templates/x11vnc.service.j2, templates/novnc.service.j2
  • 0Result and completion. Does not say what the result is
  • 0Inputs and preconditions. Does not say what the process needs to start
  • 0Progress reporting. Says nothing while it works
  • 20When it triggers. No condition that starts the skill
  • 100Steps. 53 steps
  • 100Failures and branches. 2 branches, has a failure section
  • 100Consistency. Name and required fields are in place
  • 100Execution cost. Instruction body is 3563 tokens
  • 100Running it twice. Mutating operations check current state
  • medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
  • low 16 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)
  • +3Description length 101: 120–800 characters recommended
  • +3Output format is not stated: the model decides each time
  • -4Absolute local paths (C:\Users, /home/…): not portable
  • -31 of 7 scripts are never mentioned in SKILL.md
  • +1No license
  • +2Single-language instructions
  • +4Structure: 28 headings
  • +3Step-by-step instructions: 53 items
  • +4Has examples (9 code blocks)

Quality base 70; lint remarks subtract, signals add up to 100. Result: 52.

External checks

ClawHub: suspicious
This is a disclosed remote browser automation skill, but it can expose live browser sessions and save login cookies or page data to disk with limited built-in containment.
LLM: suspicious (high) · VirusTotal: · 29 May 2026