metamask-agent-workflows
Pass
Audited by Gen Agent Trust Hub on Jul 3, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill relies on the MetaMask CLI (
mm) to perform all on-chain operations. It also executes a local Python script,scripts/amount_to_hex.py, to convert human-readable currency amounts into the hexadecimal format required by the CLI.\n- [EXTERNAL_DOWNLOADS]: The skill fetches real-time market data, reserve information, and transaction payloads from the Aave V3 GraphQL API (https://api.v3.aave.com/graphql). This is a well-known service and the data is necessary for the skill's primary functionality.\n- [PROMPT_INJECTION]: An indirect prompt injection surface exists as the skill processes data from an external API and uses it to construct command-line arguments.\n - Ingestion points: Data enters the agent's context from
https://api.v3.aave.com/graphqlviacurlcommands in files likeworkflows/aave-borrow.mdandworkflows/aave-supply.md.\n - Boundary markers: None present for the API output processing.\n
- Capability inventory: The skill can execute
mm wallet send-transactionusing arguments derived from the API data.\n - Sanitization: No explicit sanitization or validation of the GraphQL response was found in the instructions.
Audit Metadata