rlm-distill-ollama
Pass
Audited by Gen Agent Trust Hub on Apr 3, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious behavior or security policy violations were detected. The skill functions as a local developer tool for file summarization.
- [COMMAND_EXECUTION]: The skill executes Python scripts (
distiller.py,inject_summary.py,inventory.py) to facilitate repository scanning and summary generation. These operations are local and consistent with the skill's stated purpose. - [EXTERNAL_DOWNLOADS]: The skill documentation includes standard instructions for dependency management using
pip. While it references a rootrequirements.txt, the documentation states that the tool relies on the Python standard library. - [PROMPT_INJECTION]: As the skill reads and summarizes repository files, it is subject to a surface for indirect prompt injection if target files contain adversarial instructions. However, the skill does not exhibit any specific vulnerability to these injections, and the output is confined to local cache files.
- Ingestion points: Repository files read during the distillation process.
- Boundary markers: None specified in the instructions or scripts.
- Capability inventory: File reading, local file writing (JSON caches), and Python script execution via Bash.
- Sanitization: No explicit sanitization or instruction-filtering was identified in the provided file pointers.
Audit Metadata