Agentic Lending Morpho Readonly
PassAudited by ClawScan on May 7, 2026.
Overview
This appears to be a read-only Morpho lending planning skill, with the main caveats that it points to a local planner script not included here and may reuse prior run/cache artifacts.
This skill is appropriate for read-only Morpho/Api3 lending planning, not for executing transactions. Do not provide private keys or signer access. If you use the suggested Node command, run it only from a trusted repository after reviewing the referenced script, and verify whether results came from live RPC reads or cached prior artifacts.
Findings (2)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
If the referenced local script is unsafe or from the wrong repository, running it could do more than the skill intends.
The skill recommends running a local Node planner script, but this package is instruction-only and does not include that script for review. The command is purpose-aligned and described as read-only, but its actual safety depends on the external/local repository implementation.
`node bin/part2-morpho-planner.js prepare-morpho-market --input-file ./request.json`
Before running the suggested command, confirm the repository is trusted, inspect the referenced script, and verify it only performs read-only planning or verification.
Planning output could be wrong if it is based on outdated or tampered prior run data.
The workflow may rely on persistent artifacts or caches from previous runs. This is relevant to the stated planning purpose, but stale, incorrect, or untrusted artifacts could influence the agent’s analysis.
prior run artifacts can still inform read-only analysis ... funded-feed cache artifacts can inform read-only readiness analysis across runs
Use only trusted run directories and cache files, and ask the agent to distinguish live RPC reads from cached or historical artifacts.