grimoire-uniswap
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: Fetches the @grimoirelabs/cli package from the official NPM registry via npx. This is an expected behavior for CLI-based skills and targets a well-known service.
- [COMMAND_EXECUTION]: Executes the grimoire CLI and bun to perform metadata lookups, which are the primary functions of this skill. The skill also suggests checking adapter status with a doctor command.
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied inputs such as token symbols, addresses, and custom RPC/Graph endpoints which could potentially contain malicious instructions.
- Ingestion points: Command flags like --token0, --symbol, and --source in SKILL.md.
- Boundary markers: None present to delimit user input from command logic.
- Capability inventory: Executes CLI commands to fetch and display blockchain metadata.
- Sanitization: Relies on the external CLI tool for input validation and sanitization.
Audit Metadata