discovery
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill functions primarily as a documentation assistant. It creates and modifies files within a specific directory (.docs/discovery/) to capture requirements. Analysis of the instructions confirms there are no attempts to access sensitive system files, environment variables, or external network resources.
- [INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection because it writes untrusted user dialogue into a persistent file. 1. Ingestion points: User dialogue and initial instructions provided during the session. 2. Boundary markers: The skill uses a markdown template with YAML frontmatter and standard headers to structure the output. 3. Capability inventory: The skill has the ability to create and write to the local filesystem within the .docs/discovery/ path. 4. Sanitization: There is no explicit sanitization or instruction to filter user-provided content before writing to the memo.
Audit Metadata