transfer-failed-retry
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted metadata (file paths, names, and history records) retrieved via the
moviepilot_api. - Ingestion points: Data enters the context from the
transfer.historycall described in Step 1 ofSKILL.md. - Boundary markers: The skill lacks explicit boundary markers or instructions for the agent to ignore potentially malicious instructions embedded in the filenames or paths during processing.
- Capability inventory: The agent can delete history records (
transfer.history.delete) and initiate new file transfers (transfer.file), creating a path for automated actions based on external data. - Sanitization: There is no mention of sanitizing or validating the retrieved metadata before it is passed to identification tools like
media.recognizeormedia.search.
Audit Metadata