the-deal-gauge
Pass
Audited by Gen Agent Trust Hub on Aug 14, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided deal descriptions and behavioral signals that could potentially contain malicious instructions. However, the skill implements robust defensive measures by referencing comprehensive security guidelines.
- Ingestion points: The
Inputssection inSKILL.mdcaptures deal details, contact roles, and signals (e.g., website visits, content downloads). - Boundary markers: The
references/agent-security.mdfile provides strict protocols to treat all ingested content as data rather than instructions and to quote injection attempts rather than following them. - Capability inventory: The skill is authorized to read from
.agents/product-context.mdand write to.agents/gtm-run-state.md. - Sanitization: The
references/agent-security.mddocumentation mandates HTML-escaping for all interpolated values in emitted markup to prevent cross-site scripting (XSS). - [DATA_EXPOSURE]: The skill manages deal state but includes explicit restrictions against storing raw personal data or special-category information.
- Evidence: Guidelines in
SKILL.mdandreferences/agent-security.mdforbid the persistence of special-category data (e.g., race, religion, health status) and require the redaction of credentials found in inputs. - [SAFE]: No obfuscation, unauthorized network operations, or hardcoded credentials were found in the skill instructions or scripts.
Audit Metadata