release-publish
Pass
Audited by Gen Agent Trust Hub on Nov 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process Orchard release 'evidence', which is an untrusted external data source that could contain hidden instructions targeting the agent.
- Ingestion points: Orchard release evidence (referenced in SKILL.md)
- Boundary markers: Absent in the instructions
- Capability inventory: Requests network permissions (routing.json)
- Sanitization: No sanitization or validation steps are defined for the release evidence
- [NO_CODE]: The analyzed files consist only of markdown instructions and JSON metadata. No shell scripts, Python code, or Node.js logic are present within the skill folder.
Audit Metadata