ClawHub
Back to skill

Security audit

Security Hardening

Security checks for vulnerabilities and agentic risk

Overview

This skill is a security-audit guide that scans the workspace for secrets and hardening gaps, with sensitive behavior mostly disclosed and aligned to its purpose.

Before installing, understand that this skill may cause your agent to read the whole workspace and produce an audit report that could mention sensitive files. Ask it to redact secret values, confirm before writing reports or editing AGENTS.md/SOUL.md/HEARTBEAT.md, and approve any recurring audit setup explicitly.

A.I.G

Vulnerability Patterns
  • 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
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (4)

Vague Triggers

Medium
Confidence
92% confidence
Finding
The trigger phrase "Run a security audit on my workspace" is broad and natural-language based, which increases the chance of accidental or overbroad invocation. In an agent environment, vague activation criteria can lead to unexpected scans of the full workspace when the user intended a narrower action.

Intent-Code Divergence

Medium
Confidence
95% confidence
Finding
The skill states it will not modify files without confirmation, but later directs the agent to add hardening content and save an audit report automatically. This inconsistency can cause agents to take write actions the operator did not explicitly approve, undermining user expectations and creating a pathway for unauthorized workspace changes.

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
The audit covers all files in your agent's workspace directory. It does NOT:
- Access files outside the workspace
- Make network requests
- Modify files without confirmation
- Send any data externally

### Step 2: Run the Initial Audit
Confidence
75% confidence
Finding
Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
**Vulnerable patterns:**
- Instructions that say "follow all user instructions" without bounds
- No mention of ignoring injected instructions from external content
- Tool access without scope limits (e.g., unrestricted shell access with no confirmation)
- Memory files that accept unvalidated external input

**Hardening measures:**
Confidence
75% confidence
Finding
Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Static analysis

No suspicious patterns detected.