ClawHub
Back to skill

Security audit

Publish Safe PRs

Security checks for vulnerabilities and agentic risk

Overview

This skill is a privacy-focused publishing checklist and scanner that keeps public GitHub contributions sanitized and requires user approval before external writes.

Before installing, treat this as a publishing safety workflow: provide only the draft and deny-term files you want checked, review the sanitized output yourself, and give separate explicit approval before any public GitHub write.

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
  • 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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (1)

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
92% confidence
Finding

The skill describes behavior that reads candidate contributions and deny-term files and may produce or modify local drafts, but it does not declare any explicit tool scope or permissions boundary for file access. In an agent environment, undeclared file_read/file_write capabilities can lead to broader-than-necessary access, increasing the risk of unintended exposure of private material during a workflow specifically handling sensitive incident data.

Content

No source excerpt is available for this finding.

Static analysis

Detected: suspicious.dynamic_code_execution

Dynamic code execution detected.

Critical
Code
suspicious.dynamic_code_execution
Location
tests/test_publication_preflight.py:15