blueprint-docs-list
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION] (SAFE): The skill uses standard Bash utilities like head, grep, and sed to parse markdown files for metadata. These operations are performed on local files within specific documentation directories and do not involve executing untrusted input.
- [DATA_EXFILTRATION] (SAFE): There are no network-capable tools or patterns used in the skill. Data remains in the local execution environment.
- [PROMPT_INJECTION] (LOW): The skill has an indirect prompt injection surface as it reads data from potentially untrusted files in the documentation directory. 1. Ingestion points: files in docs/adrs, docs/prds, and docs/prps. 2. Boundary markers: No delimiters are used in the output table. 3. Capability inventory: Bash (local file reading). 4. Sanitization: Basic text extraction using sed.
Audit Metadata