ClawHub
Back to skill

Security audit

blast-radius

Security checks for vulnerabilities and agentic risk

Overview

This skill performs disclosed, read-only code impact analysis and has no evidence of hidden data collection, persistence, or destructive behavior.

Install only if you are comfortable with the skill checking your local Claude plugin directory for gauntlet and running that local helper when present. Without gauntlet, it falls back to repository-local git and search commands.

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
  • 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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (1)

Agent Config Directory Access

High
Category
Agent Snooping
Content
data. Check if it's available:

```bash
GRAPH_QUERY=$(find ~/.claude/plugins -name "graph_query.py" -path "*/gauntlet/*" 2>/dev/null | head -1)
```

**If gauntlet is not installed** (GRAPH_QUERY is empty):
Confidence
92% confidence
Finding
The skill instructs the agent to enumerate `~/.claude/plugins` to locate a helper script in the user's local agent configuration. Even though the command is read-only and narrowly scoped, it crosses a trust boundary by probing agent-local filesystem state outside the repository being analyzed, which can disclose installed plugins and environment layout and normalize access to sensitive config directories.

Static analysis

No suspicious patterns detected.