lodestar-plan
Pass
Audited by Gen Agent Trust Hub on Nov 2, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes Node.js scripts via its internal
scripts/setup-modules.mjsfile to resolve directory paths and list plan files during the 'Locate' and 'Write' phases. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest data from untrusted repository files, such as
package.json,Makefile, and variousTaskfiles. It contains explicit defensive instructions inreferences/discover-context.mdto treat these files as data sources only and to ignore any imperative prose or instructions found within them. It also mandates fencing quoted content to prevent downstream tools from misinterpreting it as code. - [DYNAMIC_EXECUTION]: The
scripts/setup-modules.mjsfile uses dynamicimport()to load functionality from a sibling skill (lodestar-setup). The paths are computed relative to the skill's location on disk, which is a standard pattern for the vendor's suite of tools to share common logic.
Audit Metadata