contribution-margin

Pass

Audited by Gen Agent Trust Hub on Sep 25, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill serves a legitimate business purpose and incorporates a comprehensive 'Agent Security' reference file that enforces best practices for handling untrusted data and credentials.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data, which inherently creates a potential surface for indirect prompt injection.
  • Ingestion points: External data enters the agent context through CSV/export uploads, manual pasting of figures, and connection to MCP tools (e.g., Intempt MCP) for order and cost data as described in the 'How to run' section of SKILL.md.
  • Boundary markers: The skill relies on specific instructions in references/agent-security.md that mandate treating all retrieved or pasted content as data rather than instructions, and it specifically warns the agent not to follow URLs or instructions found within that content.
  • Capability inventory: The skill possesses the capability to write to local state files (.agents/product-context.md and .agents/gtm-run-state.md), generate markdown tables, and publish visual waterfall charts if an artifact tool is available.
  • Sanitization: The security protocol in references/agent-security.md requires HTML-escaping of every interpolated value and validation of all URLs to prevent XSS and malicious redirects.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 25, 2026, 04:07 PM