messari-deep-research
Pass
Audited by Gen Agent Trust Hub on Mar 21, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill includes a bash script that executes local commands such as
mkdirfor directory creation and calls thempCLI tool to interact with the Messari API. - [EXTERNAL_DOWNLOADS]: The workflow communicates with
api.messari.io, a well-known industry data provider, to initiate research jobs and retrieve completed reports. - [PROMPT_INJECTION]: The skill processes and displays content returned from an external API, which constitutes an indirect prompt injection surface.
- Ingestion points: Data is received from the Messari API response (SKILL.md).
- Boundary markers: No explicit delimiters or safety instructions are provided to the agent to treat the fetched research content as untrusted (SKILL.md).
- Capability inventory: The skill possesses filesystem access via bash redirection and the
mkdircommand (SKILL.md). - Sanitization: No specific sanitization or validation of the external JSON data is performed before it is saved or presented (SKILL.md).
Audit Metadata