clerk
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to perform version detection by checking
package.json. This introduces a vulnerability surface where an attacker-controlled project file could theoretically contain malicious instructions. - Ingestion points:
package.json(referenced in SKILL.md). - Boundary markers: None present in the routing instructions.
- Capability inventory: None. This skill is a router and does not include any scripts, tools, or executable commands.
- Sanitization: None present.
Audit Metadata