research-project-memory
Pass
Audited by Gen Agent Trust Hub on May 10, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides a robust framework for managing research data with a focus on data integrity and privacy. It includes a dedicated 'Source Visibility Protocol' specifically designed to prevent the accidental exposure of private agent memory, raw data, or internal notes in collaborator-visible or public repositories.
- [SAFE]: The instructions emphasize the separation of stable project facts from volatile runtime state and provide clear guidance on maintaining provenance (traceability) for all research claims and results.
- [SAFE]: All shell commands described in the protocols are standard development tools used for version control (git), project management (gh), environment management (uv), and code quality (ruff, mypy, pytest, tex-fmt). These are provided as templates for user configuration rather than being executed automatically or maliciously.
- [SAFE]: The skill implements a 'least privilege' approach by instructing agents to keep private paths and credentials out of shared project memory and to re-verify volatile facts before acting.
Audit Metadata