coverage-analyzer
Fail
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: HIGHPROMPT_INJECTION
Full Analysis
- [Indirect Prompt Injection] (HIGH): The skill architecture allows external data to influence side-effect operations.
- Ingestion points: Ingests local source code through Read, Grep, and Glob tools.
- Boundary markers: Absent; no delimiters are defined to isolate untrusted code content from agent instructions.
- Capability inventory: Possesses Write and Edit permissions, allowing the agent to persist changes in the local environment.
- Sanitization: Absent; the skill lacks input validation or output sanitization to filter malicious instructions embedded in data.
Recommendations
- AI detected serious security threats
Audit Metadata