The Agent Skills Directory

[SAFE]: The skill implements a deterministic strategy reviewer in Python. No malicious patterns or security risks were identified.
[DATA_EXPOSURE_AND_EXFILTRATION]: The script reads draft strategy files and writes reports to local directories. It does not perform network operations, use credentials, or access sensitive system locations (e.g., .ssh, .aws).
[DYNAMIC_EXECUTION]: The tool uses yaml.safe_load() in scripts/review_strategy_drafts.py to parse draft files, which is a best practice that prevents arbitrary code execution during the YAML parsing process.
[INDIRECT_PROMPT_INJECTION]: The skill processes external YAML drafts which represent an untrusted data surface.
Ingestion points: scripts/review_strategy_drafts.py reads YAML files from a user-provided directory.
Boundary markers: None; the tool parses and evaluates fields directly.
Capability inventory: The script is limited to local file system reads and writes; it lacks network access or shell execution capabilities.
Sanitization: The script uses yaml.safe_load and enforces strict type casting (float, int) for the values it evaluates, mitigating risks associated with malicious data injection.

edge-strategy-reviewer