dev-context-multi-repo

Pass

Audited by Gen Agent Trust Hub on May 20, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [SAFE]: The skill implements a robust framework for managing context across multiple code repositories. It performs passive analysis of local source code files and project metadata to generate structured documentation and relationship graphs. All file operations are confined to the user-specified root directories and the skill's own scaffolded directories.
  • [COMMAND_EXECUTION]: Several scripts, including scripts/build_artifact_set.py, scripts/scan_portfolio.py, and scripts/incremental_update.py, use the subprocess module to execute internal toolchain commands. These executions are restricted to calling either the skill's own Python scripts (using sys.executable) or the local git binary to retrieve commit metadata. This is standard functional behavior for a multi-repo orchestration tool and is implemented using deterministic paths or user-supplied repository roots.
Audit Metadata
Risk Level
SAFE
Analyzed
May 20, 2026, 07:48 AM
Security Audit — agent-trust-hub — dev-context-multi-repo