loop-me
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from the local workspace, which represents a potential surface for indirect prompt injection if those files are modified by external actors.
- Ingestion points: The agent is instructed to read context and terminology from
NOTES.mdand existing workflow specifications inworkflows/*.md. - Boundary markers: The instructions do not define specific delimiters or "ignore previous instruction" markers when reading these files to prevent the agent from executing commands embedded in the text.
- Capability inventory: The skill possesses the ability to create, edit, and delete files within the
workflows/directory and modifyNOTES.mdbased on its interactions. - Sanitization: There is no evidence of content validation or escaping mechanisms for the data ingested from the workspace files.
Audit Metadata