grimoire-pendle

Pass

Audited by Gen Agent Trust Hub on Oct 16, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructions suggest using npx -y @grimoirelabs/cli and @grimoirelabs/venues, which triggers the download and execution of packages from the NPM registry at runtime.
  • [INDIRECT_PROMPT_INJECTION]: The skill operates by ingesting metadata (market addresses, token details) from the Pendle Finance API and using it to construct transaction 'spells'. This creates an attack surface where malicious or unexpected API responses could influence the generated code.
  • Ingestion points: Data is fetched from api-v2.pendle.finance via CLI commands defined in SKILL.md.
  • Boundary markers: No explicit delimiters are specified to separate untrusted API data from the agent's instructions.
  • Capability inventory: The skill is designed to generate and potentially execute blockchain transactions (e.g., pendle.swap, pendle.add_liquidity).
  • Sanitization: The instructions do not mention validation or sanitization steps for the data retrieved from the Pendle API.
  • [COMMAND_EXECUTION]: The skill relies on executing the grimoire CLI and its subcommands to interact with the Pendle protocol and local environment.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 16, 2026, 11:09 PM