read-all-adrs
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONPROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to read all files in the docs/adr/ directory, which can introduce malicious instructions into the agent's context if the files are compromised. 1. Ingestion points: Markdown files in the docs/adr/ folder (SKILL.md). 2. Boundary markers: Absent. 3. Capability inventory: File reading and processing capabilities. 4. Sanitization: Absent.
- [PROMPT_INJECTION]: The skill uses aggressive, imperative language ('DO NOT BE A LAZY CUNT. ACTUALLY DO THIS.') designed to override the agent's default behavior or tendency to summarize content.
Audit Metadata