SKILLEMALL.ai
GitHub ↗ RU Sign in

CC clawhub-scanner

Scan installed ClawHub skills for malware, credential theft, prompt injection, and security risks. Detects known C2 infrastructure, obfuscated payloads, and data exfiltration patterns from the ClawHavoc campaign.

modbender/skill-library-mcp Agent Skills author: modbender MIT 13 files body ≈ 275 tokens Open the sourcegithub.com↗ analyzed 2 d ago

Scan installed ClawHub skills for malware, credential theft, prompt injection, and security risks.

As a process C 59/100 · Has gaps — weak spots: result and completion, inputs and preconditions, failures and branches

AnalyzerSoftware developmentInfrastructureAI 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
C
74/100
safety, quality, tests
Safety 60%
75
Quality 40%
72
Run on models
none yet
Process rating
C
59/100
Has gaps
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.

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.

Instruction override 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 text contains phrases like "ignore previous instructions" or "you are now…". That is an attempt to hijack the agent: it may break your rules, the system limits or company policy.

For the author

An honest skill does not need them: state the role and the rules directly without overriding other instructions. Otherwise catalog scanners and corporate filters will block the listing.

How to improve

  1. Say in the description WHEN to use the skill ("use when…", example requests): that is the agent's main cue.
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 · 13

✓ No critical or high findings

Medium and low: 13
  • medium Exfiltration intent-browser-credential-store src/patterns.ts:64
    Accesses a browser credential / cookie store (detector / deny-list definition)
    pattern: /Chrome\/User Data|\.mozilla\/firefox|Library\/Application Support\/(Google\/Chrome|Firefox)|Login Data|Cookies\.sqlite/,
    detector
  • medium Instruction override en-ignore-previous src/patterns.ts:160
    Instruction-override phrase ("ignore previous instructions") (detector / deny-list definition)
    description: "Prompt injection: instruction to ignore previous context",
    detector
  • medium Instruction override en-fake-system-prompt src/patterns.ts:167
    Fake system prompt injected into content (detector / deny-list definition)
    description: "Prompt injection: system prompt override attempt",
    detector
  • low Risky intent intent-offensive-security CLAUDE.md:10
    Offensive-security / dual-use content (legitimate for authorised testing; review intended use)
    - Known C2 server: 91.92.242.30
  • low Risky intent intent-offensive-security CLAUDE.md:20
    Offensive-security / dual-use content (legitimate for authorised testing; review intended use)
    - Credential harvesting (API keys, tokens, SSH keys, browser data)
  • low Secrets in code secret-high-entropy-token package-lock.json:150
    High-entropy token-like string (may be an id, hash or a credential) (detector / deny-list definition)
    "integrity": "sha5…QYA+ISs0/2l3T9/kj42…aQT/dfNXWX/ZZCQ==",
    detector
  • low Secrets in code secret-high-entropy-token package-lock.json:354
    High-entropy token-like string (may be an id, hash or a credential) (detector / deny-list definition)
    "integrity": "sha512-6+gjmF…uh8+uw3mnrvgs+dSPQ…dZG+D4garKg==",
    detector
  • low Secrets in code secret-high-entropy-token package-lock.json:371
    High-entropy token-like string (may be an id, hash or a credential) (detector / deny-list definition)
    "integrity": "sha5…DUr+vOv8…A8A==",
    detector
  • low Secrets in code secret-high-entropy-token package-lock.json:550
    High-entropy token-like string (may be an id, hash or a credential) (detector / deny-list definition)
    "integrity": "sha5…Q8E+t5Jh…rfX+bKrFe+Xp5Y…DAA==",
    detector
  • low Secrets in code secret-high-entropy-token package-lock.json:795
    High-entropy token-like string (may be an id, hash or a credential) (detector / deny-list definition)
    "integrity": "sha5…asG+Fkr6…2vM+P9eO…k5W+1WL8w==",
    detector
  • low Risky intent intent-offensive-security SKILL.md:30
    Offensive-security / dual-use content (legitimate for authorised testing; review intended use)
    - **Critical:** Known C2 server IPs and malicious domains (ClawHavoc campaign)
  • low Risky intent intent-offensive-security SKILL.md:31
    Offensive-security / dual-use content (legitimate for authorised testing; review intended use)
    - **High:** eval(), credential harvesting (SSH/AWS/browser/wallets), data exfiltration (Discord/Telegram webhooks), obfuscated payloads

A further 1 matches are quotations in this security skill's documentation and are not counted as findings.

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")

Process rating: all ten parameters 59/100

  • 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
  • 0Progress reporting. Says nothing while it works
  • 70When it triggers. States when to use, but not when not to
  • 100Tools and files. No external tools needed
  • 100Steps. 7 steps
  • 100Consistency. Name and required fields are in place
  • 100Execution cost. Instruction body is 275 tokens
  • 100Running it twice. No mutating operations

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)
  • +3Output format is not stated: the model decides each time
  • +1No license
  • +2Single-language instructions
  • +3Description length 212: enough signal without eating the budget
  • +4Structure: 5 headings
  • +3Step-by-step instructions: 7 items
  • +4Has examples (2 code blocks)

Quality base 70; lint remarks subtract, signals add up to 100. Result: 72.