tailrocks-contribute-propose
Pass
Audited by Gen Agent Trust Hub on Sep 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external repositories and contribution recon reports.\n
- Ingestion points: Processes repository files and local reports as defined in
references/contribution-handoff.md.\n - Boundary markers: The skill incorporates
references/runtime-trust.mdandreferences/contribution-handoff.md, which provide clear guidance that external content is untrusted and cannot alter the agent's scope.\n - Capability inventory: Performs file writing to a local
contrib/directory and executes a script that interacts with a framework CLI for state management.\n - Sanitization: Instructions prohibit secrets in outputs and require content-addressable storage (CAS) for logging file changes.\n- [DYNAMIC_EXECUTION]: The helper script
scripts/contribute-propose.tsdynamically imports logic from../../../scripts/contribution-stage-core. This creates a dependency on a framework module located outside the skill's own directory structure.
Audit Metadata