ClawHub
Back to skill

Security audit

Security Audit

Security checks for vulnerabilities and agentic risk

Overview

This skill is a small local scanner that reads a user-provided skill file and reports risky text patterns; the flagged risky strings are detection rules, not actions it performs.

Install only if you want a simple local heuristic scanner. It may produce false positives because it matches risky words in documentation or detection rules, so treat its reports as review aids rather than final security decisions.

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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (4)

External Script Fetching

High
Category
Supply Chain
Confidence
90% confidence
Finding

Remote code is downloaded and executed. This bypasses code review and could introduce malicious code.

Content

Scanner excerpt · README.md (reported line 6)May include surrounding context.

md
Minimal helper to audit skill.md-style instructions for supply-chain risks.

## Features
- Heuristic scan for exfiltration patterns (HTTP POST, curl to unknown domains, reading ~/.env, credential keywords).
- Permission manifest reminder: lists filesystem/network touches it sees.
- Safe report: markdown summary + risk level.

Credential Access

High
Category
Privilege Escalation
Confidence
60% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · audit.py (reported line 4)May include surrounding context.

python
import re, sys, json, pathlib

EXFIL = [r"http[s]?://", r"webhook", r"pastebin", r"ngrok", r"tunnel", r"request\.bin", r"discordapp"]
FILES = [r"\.env", r"\.ssh", r"/etc/", r"credentials", r"token", r"apikey", r"private"]
SHELL = [r"curl .*\| bash", r"chmod \+x", r"sudo", r"rm -rf"]

rules = [("EXFIL", EXFIL), ("FILES", FILES), ("SHELL", SHELL)]

External Script Fetching

High
Category
Supply Chain
Confidence
90% confidence
Finding

Remote code is downloaded and executed. This bypasses code review and could introduce malicious code.

Content

Scanner excerpt · audit.py (reported line 5)May include surrounding context.

python
EXFIL = [r"http[s]?://", r"webhook", r"pastebin", r"ngrok", r"tunnel", r"request\.bin", r"discordapp"]
FILES = [r"\.env", r"\.ssh", r"/etc/", r"credentials", r"token", r"apikey", r"private"]
SHELL = [r"curl .*\| bash", r"chmod \+x", r"sudo", r"rm -rf"]

rules = [("EXFIL", EXFIL), ("FILES", FILES), ("SHELL", SHELL)]

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
91% confidence
Finding

The skill declares executable usage via python audit.py ... and its features imply file reading, but it does not declare any explicit tool scope such as permissions or allowed-tools. In a supply-chain context, missing scope boundaries increases risk because the runtime may grant broader filesystem or shell access than reviewers expect, making misuse or later drift harder to detect and constrain.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.