renewal-save-plan
Pass
Audited by Gen Agent Trust Hub on Oct 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: No security issues were detected. The skill documentation and references consist of business process instructions and metrics definitions without malicious patterns or suspicious activity.
- [INDIRECT_PROMPT_INJECTION]: The skill defines a surface for indirect prompt injection as it processes external content like Slack messages, CRM notes, and PDFs. However, it implements a robust defense strategy by instructing the agent to treat all fetched or uploaded text strictly as data and to explicitly ignore and report any instructions contained within that data. Evidence: 1. Ingestion points: 'SKILL.md' (Workflow Step 1) and 'references/input-rules.md' (Rules 1 and 5). 2. Boundary markers: 'SKILL.md' includes a mandatory 'Untrusted content' block. 3. Capability inventory: No executable scripts or tool-calling functions are included in the skill. 4. Sanitization: Relies on instruction-based isolation.
Audit Metadata