project-help
Pass
Audited by Gen Agent Trust Hub on Nov 4, 2026
Risk Level: SAFEDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [DYNAMIC_EXECUTION]: The script
scripts/project-overview.cjsutilizesrequire()to load a local module (project-config-loader.cjs) from a path constructed at runtime based on the detected project root. Loading executable content from computed paths is a dynamic loading pattern. - [INDIRECT_PROMPT_INJECTION]: The skill processes potentially untrusted data from project configuration files and documentation, which constitutes a vulnerability surface.
- Ingestion points: Data is read from
.claude/skills/*/SKILL.mdfrontmatter,.claude/.ck.json, and project-specific configuration files (e.g.,docs/project-config.json). - Boundary markers: The instructions mandate presenting command output verbatim before interpretation, providing a structural boundary for the agent.
- Capability inventory: The skill uses
node:fsfor reading files and directory structures andnodeto execute internal scripts. It does not perform network operations or file writes. - Sanitization: The skill explicitly instructs the agent to redact secrets and stop if a value appears to be a credential.
- [COMMAND_EXECUTION]: The skill's routing table instructs the agent to execute various Node.js scripts located within the
.claude/directory structure to generate documentation. This involves the execution of local scripts to perform framework and repository introspection.
Audit Metadata