grimoire-morpho-blue
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill establishes an indirect prompt injection surface by consuming external protocol metadata to influence agent-authored financial transactions.
- Ingestion points: Market metadata and vault snapshots fetched via the
vaults-snapshotcommand inSKILL.md. - Boundary markers: No instructions or delimiters are provided to the agent to treat the fetched metadata as untrusted data when interpolating it into parameters.
- Capability inventory: The skill provides templates for high-value financial actions including
morpho_blue.lend,morpho_blue.borrow,morpho_blue.supply_collateral, andmorpho_blue.vault_deposit. - Sanitization: There is no evidence of validation or sanitization of the externally sourced market data before it is used to construct transaction sequences.
- [COMMAND_EXECUTION]: The skill relies on the execution of shell commands through the
grimoireCLI to retrieve protocol metadata and check liquidity. - [EXTERNAL_DOWNLOADS]: The skill facilitates the download and execution of external code from a public registry. It recommends the use of
npx -y @grimoirelabs/cli, which fetches the package from the NPM registry at runtime. While the registry is a well-known service, the package originates from an unverified third-party account. - [REMOTE_CODE_EXECUTION]: The recommendation to execute unverified packages via
npxfrom a public registry constitutes a remote code execution pattern.
Audit Metadata