release-director
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and follow instructions from an external configuration file named
release-preferences.mdto modify the release workflow. - Ingestion points: The
SKILL.mdfile explicitly instructs the agent toload_override("release-preferences.md")and incorporate its content into the operational logic. - Boundary markers: There are no boundary markers or instructions for the agent to ignore potentially malicious commands within the override file.
- Capability inventory: The agent has access to powerful tools including
Bash,Write,Edit, and the vendor-providedbitwize-music-mcptoolset, which could be misused if the agent is successfully injected. - Sanitization: No sanitization or validation logic is defined for the content retrieved from the override file.
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to extract the album name from the project's
READMEfile to generate a social media announcement link. This creates an attack surface where malicious content embedded in theREADMEcould influence the agent's behavior during the text extraction and processing phase.
Audit Metadata