linkedin-engagement
Pass
Audited by Gen Agent Trust Hub on Aug 28, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill uses Python scripts (
scripts/comment_target_planner.py,scripts/outreach_message_builder.py,scripts/outreach_volume_guard.py) that rely exclusively on the Python standard library (argparse, json, sys, re). No external dependencies or network operations were detected. - [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided content (e.g., connection request reasons and specific lines) to assemble outreach messages. While this creates a potential surface for indirect prompt injection if the user provides malicious text, the scripts are designed to return text to the user for manual review and do not execute any commands derived from the input data.
- Ingestion points: User-provided strings via CLI arguments or JSON input to
scripts/outreach_message_builder.py. - Boundary markers: None explicitly used in the assembled message output.
- Capability inventory: No subprocess calls, file-writes, or network operations are present in any script.
- Sanitization: The scripts perform validation against a list of 'dead phrases' and pitch language, providing feedback to the user to improve content quality.
- [DATA_EXPOSURE_&_EXFILTRATION]: No sensitive file paths, environment variable access, or credential harvesting patterns were found. The skill explicitly states in its documentation that it does not use credentials or API calls.
Audit Metadata