progress-tracker
Pass
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: LOWPROMPT_INJECTIONNO_CODE
Full Analysis
- [NO_CODE] (INFO): Only the SKILL.md file was provided for analysis. The Python scripts mentioned in the documentation (e.g., progress_tracker.py) are missing from the file bundle.
- [PROMPT_INJECTION] (LOW): The skill presents an indirect prompt injection surface as it ingests untrusted project data to drive agent logic.
- Ingestion points: Project files within 'project_dir' are processed to calculate metrics.
- Boundary markers: No delimiters or instructions to ignore embedded commands are documented.
- Capability inventory: The 'autonomous-loop' uses generated metrics for continuation decisions, creating a surface for internal logic manipulation.
- Sanitization: No validation or sanitization of project data is described in the documentation.
Audit Metadata