competitive-win-loss-analysis
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill follows a rigorous Product Marketing methodology designed to provide objective analysis while protecting sensitive data. It explicitly instructs the agent to preserve account privacy and distinguish between factual observations and interpretations. It also emphasizes the use of primary source evidence (buyer quotes) over internal rep notes to ensure accuracy.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze external data such as call transcripts, CRM notes, and win/loss interviews. This processing of untrusted content represents a potential indirect prompt injection surface. 1. Ingestion points: CRM opportunity records, call transcripts, and interview documents referenced in references/INTERNAL.md and SKILL.md. 2. Boundary markers: The use of win-loss-template.md and separate columns for buyer-stated vs. rep-entered reasons acts as a logical delimiter for processed data. 3. Capability inventory: Tool-based access to connected CRM data, call intelligence systems, and the LMTY competitive intelligence platform for data retrieval. 4. Sanitization: The methodology mandates factual reporting, source attribution, and verbatim quoting which provides context but does not implement active character-level filtering.
Audit Metadata