win-loss-analysis

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill's primary function is to process external data sources including CRM stage history, seller notes, and interview transcripts, which introduces a surface for indirect prompt injection if the source data contains adversarial instructions.
  • Ingestion points: Data provided via CRM exports and interview text files specified in SKILL.md.
  • Boundary markers: The process uses structured markdown templates (templates/output-template.md) to organize data, but does not provide explicit instructions for the agent to ignore instructions embedded within input text.
  • Capability inventory: The skill includes a local validation script (scripts/check-output.py) for file structure verification and relies on standard file operations.
  • Sanitization: There are no explicit filtering or sanitization steps defined for external data input.
  • [SAFE]: References industry-standard research and framework documentation from clozd.com, productmarketingalliance.com, and jobstobedone.org.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 12:46 PM
Security Audit — agent-trust-hub — win-loss-analysis