ClawHub
Back to skill

Security audit

CCPA Audit - CCPA合规审计工具

Security checks for vulnerabilities and agentic risk

Overview

The skill is mostly a disclosed cloud-scored CCPA audit tool, but its package privacy statement understates default answer submission and the HTML report renderer trusts cloud-returned content.

Review before installing. Use this only if you are comfortable sending pass/fail CCPA audit answers to compliancehub.cn, including in anonymous trial mode. Prefer setting an API key deliberately, delete ~/.config/compliancehub/ccpa-audit.anon_id if you do not want trial linkability, and avoid opening HTML reports from untrusted or unexpected cloud responses until the renderer escapes dynamic fields.

A.I.G

Vulnerability Patterns
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
Findings (2)

T09 · Insecure Skill Coding Practices

Warning
Location
scripts/ccpa-audit.py:329
Finding

Unescaped Cloud-Controlled Data in Generated HTML Reports

Content
View full analysis
= 80 else "#ff9800" if score >= 60 else "#f44336" rows = "" current_cat = "" for r in items: if r.get("category_name") != current_cat: current_cat = r.get("category_name", "") rows += f'{current_cat}\n' icon = "✅" if r.get("passed") else "❌" cls = "pass" if r.get("passed") else "fail" rec = r.get("recommendation") or "Keep it up" rows += f"""{icon}{r.get('name')}{r.get('legal_ref') or ''}{cls.upper()}{rec}\n""" return f""" CCPA Compliance Audit Report body{{font-family:-apple-system,BlinkMacSystemFont,'Segoe UI',sans-serif;max-width:960px;margin:2rem auto;padding:0 1rem;color:#333}} h1{{border-bottom:2px solid #2563eb;padding-bottom:.5rem}} .score-card{{background:linear-gradient(135deg,#2563eb,#1d4ed8);color:#fff;padding:2rem;border-radius:12px;text-align:center;margin:1.5rem 0}} .score{{font-size:4rem;font-weight:700}} .summary{{display:flex;gap:2rem;justify-content:center;margin-top:1rem}} .summary div{{text-align:center;font-size:1.2rem}} table{{width:100%;border-collapse:collapse;margin-top:1.5rem}} th{{background:#f1f5f9;text-align:left;padding:.75rem;border-bottom:2px solid #e2e8f0}} td{{padding:.75rem;border-bottom:1px solid #e2e8f0}} tr.pass td:first-child{{color:#4caf50}} tr.fail td:first-child{{color:#f44336}} tr.category-row td{{background:#dbeafe;font-weight:600;color:#1d4ed8}} .note{{color:#94a3b8;margin-top:2rem;font-size:.85rem}}

CCPA Compli ...[truncated 2562 chars]

Remediation
View remediation
``` 5. Add regression tests using hostile values such as: ```text "> ``` The tests should verify that these values appear as encoded text and cannot create executable elements. ]]>

other

Warning
Location
package.json:11
Finding

Package Privacy Declaration Contradicts Automatic Anonymous Cloud Submission

Content
View full analysis
Remediation
View remediation

SkillSpector

Vulnerability Patterns
  • Rogue AgentSelf-Modification, Session Persistence
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (4)

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 73)May include surrounding context.

md
python3 scripts/ccpa-audit.py --non-interactive

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 86)May include surrounding context.

md
python3 scripts/ccpa-audit.py --non-interactive

Session Persistence

Medium
Category
Rogue Agent
Confidence
83% confidence
Finding

The skill explicitly stores persistent identifiers and an API key in the user's home directory and relies on them across sessions, creating session persistence and credential residue on disk. While this appears to be disclosed and functionally necessary for the cloud-scored workflow, persistent tokens can be recovered by other local processes, reused unintentionally, or increase the blast radius if the host is compromised.

Content

Scanner excerpt · SKILL.md (reported line 102)May include surrounding context.

md
`https://compliancehub.cn` (the operator's official endpoint). Scoring transmits only your item answers and the free API Key (as a Bearer token).
- **API Key (you provide it):** obtained from the web account center and supplied via the
  `COMPLIANCEHUB_API_KEY` environment variable or `~/.config/compliancehub/ccpa-audit.key` (0600, outside the skill
  folder). The skill reads the Key; it does not write accounts or collect credentials.
- **No shell execution:** stdlib only (`urllib`, `json`, `ssl`); no shell, no external binaries.
- **Not a rogue/autonomous agent:** The stored files are an ordinary API key file (0600) and a random
  anonymous-trial anon_id (`~/.config/compliancehub/ccpa-audit.anon_id`, 0600, no personal data) — not

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
77% confidence
Finding

The natural-language guidance says users may directly enter Chinese answers and much of the skill text is bilingual, but it does not clearly state that English-only interaction is equally supported at runtime. This can be read as steering users toward a specific language without explicit opt-in.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.