messari-token-research
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses shell commands and a bash script to automate data retrieval and store research results in the user's local directory at
~/.config/moonpay/research/. This is standard behavior for the intended workflow. - [INDIRECT_PROMPT_INJECTION]: The skill represents a surface for indirect prompt injection as it fetches external content from the Messari API (asset details, news, signals) and interpolates it into a prompt for AI synthesis in Step 5.
- Ingestion points: Data is retrieved from
api.messari.io(Steps 1-4) and stored in local JSON files. - Boundary markers: The AI prompt uses standard role separation (
systemvsuser) but does not include specific delimiters or instructions to ignore embedded commands within the fetched data. - Capability inventory: The skill has the capability to execute shell commands (
bash) and write to the local file system (mkdir,echoto file). - Sanitization: There is no explicit sanitization or filtering of the content retrieved from the Messari API before it is passed to the AI synthesis step.
Audit Metadata