slack-proactive-intervention-patterns
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- [SAFE] (SAFE): The analysis of
references/pattern-detection-guide.mdandreferences/scoring-examples.mdshows they contain only markdown-based instructional content and examples for evaluating chat messages. No executable scripts, shell commands, or external dependencies are present. - [Indirect Prompt Injection] (SAFE): The documentation describes a system designed to process user-generated chat content (e.g., Slack messages). While this represents a potential attack surface for indirect prompt injection, these specific files do not contain the implementation logic or prompt templates that would be vulnerable. No evidence of unsafe interpolation or missing sanitization was found in these reference documents.
Audit Metadata