archive
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill reads content from feature files and design documents to sync them into living documentation, representing a surface where hidden instructions could influence agent behavior.
- Ingestion points: The skill ingests data from
beat/changes/<name>/features/*.feature,proposal.md,design.md,tasks.md, andbeat/CONTEXT.md. - Capability inventory: The skill uses shell commands for file system operations (
mkdir,mv,rm) and Git repository management (git add,git commit,git grep). - Boundary markers: There are no explicit instructions to use boundary markers or ignore embedded instructions when reading these files.
- Sanitization: The skill does not perform sanitization or validation of the text content extracted from the repository files before processing.
- [COMMAND_EXECUTION]: The skill executes shell commands for file management and Git tasks, incorporating user-provided change names and paths into command strings.
Audit Metadata