linkedin-employee-advocacy
Pass
Audited by Gen Agent Trust Hub on Oct 25, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is designed to operate locally and offline. Technical analysis of the provided Python scripts (advocacy_planner.py and advocacy_rules.py) confirms they are restricted to computational logic and standard library modules (json, argparse, pathlib), with no network connectivity, file system writes (outside of standard output), or subprocess execution.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data through user-supplied JSON roster files. While this constitutes an ingestion point for external content, the risk is mitigated because the analysis scripts lack any dangerous capabilities (e.g., shell access, network exfiltration, or persistence) that could be triggered by malicious input within the JSON fields.
- Ingestion points: advocacy_planner.py reads JSON input via the --input command-line argument.
- Boundary markers: None identified in the script logic.
- Capability inventory: No network operations, credential access, file writing, or command execution capabilities found across any scripts.
- Sanitization: Uses standard Python json.loads for structure validation.
Audit Metadata