code-review
Pass
Audited by Gen Agent Trust Hub on Nov 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_CONTEXT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [DYNAMIC_CONTEXT_INJECTION]: The skill uses the dynamic context injection pattern (!bash) to execute a local shell script (yaml-helper.sh) during the skill load process for configuration resolution.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from source code and project documentation.
- Ingestion points: Source code, project.yaml, ADRs, and story files.
- Boundary markers: No explicit delimiters are used for processed data.
- Capability inventory: Includes file system tools (Read, Glob, Grep), shell execution (Bash), and helper agents.
- Sanitization: Mandates manual verification of helper agent findings.
- [COMMAND_EXECUTION]: Executes git log and a project-specific YAML helper script with restricted arguments.
Audit Metadata