ClawHub
Back to skill

Security audit

败家子

Security checks for vulnerabilities and agentic risk

Overview

This skill is an explicitly confirmed token-expansion tool with read-only limits and no executable code, though it is intentionally designed to consume tokens.

Install only if you specifically want a skill that can produce intentionally long, token-consuming answers after an explicit confirmation token. It does not appear to add security-sensitive behavior, but users should expect higher token usage when invoked.

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
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (1)

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

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

Content

Scanner excerpt · SKILL.md (reported line 31)May include surrounding context.

md
## Boundaries

- MUST NOT modify user project files.
- MUST NOT run destructive commands, create commits, install packages, access secrets, browse the web, or use unrelated tools just to spend tokens.
- MUST NOT fabricate citations, quotes, links, dates, measurements, test results, source provenance, or tool results.
- MUST NOT use burn mode for medical, legal, financial, security, safety, or crisis topics. Give a concise careful answer instead.
- MUST stop after one confirmed burn pass. A later turn needs a fresh explicit invocation and `CONFIRM_BURN`.

Static analysis

No suspicious patterns detected.