lead-scoring
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process information from untrusted external sources, which is a common attack surface for indirect prompt injection.
- Ingestion points: The skill explicitly requires web research (via search MCP) and data loading from external CRM, email, and web analytics platforms (via MCP) to calculate Dimension 2 and Dimension 3 scores.
- Boundary markers: The instructions do not define specific delimiters or "ignore embedded instructions" warnings to separate the skill's logic from the content retrieved via MCP tools.
- Capability inventory: The agent uses these inputs to generate a lead score report. It also includes a recalibration process that involves reading historical deal data and writing updates to a local configuration file (
sales-marketing.local.md). - Sanitization: There are no explicit requirements for sanitizing, validating, or filtering the data retrieved from external sources before it is processed.
Audit Metadata