agent-mongo
Pass
Audited by Gen Agent Trust Hub on Sep 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The agent ingests data from external MongoDB databases through the
agent-mongotool. This data is untrusted and could contain instructions designed to influence agent behavior. - Ingestion points: Data is retrieved from the database via commands like
query find,query get, andquery aggregatedefined inSKILL.md. - Boundary markers: The tool outputs data in NDJSON/JSON format as described in
references/output.md, providing a structured boundary between data and instructions. - Capability inventory: The agent has access to the
agent-mongobinary viaBash, along with file system utilities (Read,Grep,Glob) as specified in the YAML frontmatter. - Sanitization: The tool implements automatic truncation for string fields exceeding 200 characters (configurable via
truncation.maxLength), which limits the payload size for potential injections.
Audit Metadata