recoup-roster-add-artist
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection by ingesting untrusted data from external sources and using it to perform writes and API calls.
- Ingestion points: Step 2 fetches Spotify search results (
GET /api/spotify/search), Step 4 fetches research profiles and web data (/research/lookup,/research/web), and Step 6 performs general web searches for social media links. - Boundary markers: The instructions do not define delimiters or specific "ignore embedded instructions" warnings when processing the data returned from these external sources.
- Capability inventory: The skill performs multiple file-write operations (creating
RELEASE.md,artist.md,audience.md, and updatingRECOUP.md) and state-modifying API calls (POST /api/artists,PATCH /api/artists/{id}) using therecoup-platform-api-accesstool. - Sanitization: There is no evidence of sanitization, escaping, or validation logic for the external content before it is synthesized into the knowledge base or used in platform API updates.
- Risk: Maliciously crafted content in a Spotify bio or a searched web page could attempt to influence the agent's behavior during the synthesis step (Step 8) or the API update steps.
Audit Metadata