retention-strategy
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze external data sources such as churn reports and market research, which creates a potential surface for indirect prompt injection.
- Ingestion points: Processes user-provided datasets (cohort retention, churn reasons) and external research findings as described in SKILL.md and references/skill-spec.json.
- Boundary markers: The skill requires the explicit classification of material claims (fact, inference, assumption, unknown) and uses an 'evidence-gate' hook to validate inputs.
- Capability inventory: Includes reasoning over provided context, web research capabilities, and the ability to trigger orchestration scripts.
- Sanitization: Implements an 'evidence-gate' hook and strict 'evidence discipline' protocols to ensure data integrity.
- [DYNAMIC_EXECUTION]: The skill instructions direct the agent to utilize specific local Python scripts for workflow routing and dependency management.
- Evidence: References to '../../scripts/dynamic_router.py' and '../../scripts/neural_router.py' are found in SKILL.md and references/skill-spec.json for building DAGs and specialist selection.
Audit Metadata