sales-pipeline-hygiene
Pass
Audited by Gen Agent Trust Hub on Sep 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted CRM data exports. Ingestion points: Workflow Step 2 (SKILL.md) describes extracting deal and activity data from external exports. Boundary markers: No specific boundary markers are defined to isolate untrusted data. Capability inventory: Analytical processing and reporting; no high-risk shell commands or network exfiltration patterns were identified. Sanitization: Includes a workflow step for cleaning and standardizing data to remove test records and normalize values.
- [NO_CODE]: The skill consists entirely of instructional markdown and reference files with no executable code or script dependencies.
- [SAFE]: The skill follows security best practices by recommending data classification rather than deletion and emphasizes manual review by managers instead of automated deal closure.
Audit Metadata