external-gitcode-ascend-megatron-migration-generator

Pass

Audited by Gen Agent Trust Hub on Apr 18, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill workflow involves executing local Python scripts (synthesize_full_patch_series.py and render_migration_artifacts.py) to process JSON-formatted impact reports. These scripts generate multiple markdown and patch files within the project directory. The execution is scoped to the provided scripts and does not involve arbitrary or externally-sourced commands.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests 'impact reports' (JSON) which represents a potential surface for indirect prompt injection. Mandatory Evidence Chain: 1. Ingestion points: impact_report.json (processed by scripts and read by the agent). 2. Boundary markers: None explicitly defined in the instructions for data ingestion. 3. Capability inventory: Local workspace modification (via 'apply' mode) and git commit creation (via 'commit' mode). 4. Sanitization: None detected; the skill relies on static script-based transformation. This finding is classified as low risk because the data processing is focused on structural transformation for developer review rather than executing behavioral instructions from the input data.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 18, 2026, 03:04 AM
Security Audit — agent-trust-hub — external-gitcode-ascend-megatron-migration-generator