ai-vendor-evaluation

Evaluate AI vendors, tools, platforms, and agent systems for executive decision support. Use when Codex is asked to compare AI vendors, draft vendor evaluation matrices, assess AI platform fit, prepar

SKILL.md

GitHub

AI Vendor Evaluation

Core Workflow

  1. Define use case, buyer, users, data sensitivity, deployment context, budget, integration needs, and risk tier.
  2. Compare vendors by capability fit, data handling, security, governance, admin controls, interoperability, cost, support, maturity, and exit risk.
  3. Separate vendor claims from verified evidence and open questions.
  4. Draft evaluation matrix, procurement questions, and pilot requirements.
  5. Identify legal, security, privacy, finance, procurement, and IT review needs.
  6. Recommend next diligence steps, not final procurement approval.

Safety Rules

  • Do not claim a vendor is compliant, secure, approved, or best without current evidence and owner review.
  • Verify current official vendor documentation before platform-specific claims.
  • Do not recommend sharing sensitive data with a vendor without approval.
  • Escalate procurement, contract, data processing, security, privacy, employment, customer, and regulated-use risks.

Deliverable Shape

For AI vendor evaluation, provide:

  • Evaluation goal and scope
  • Vendor comparison matrix
  • Evidence and open questions
  • Security and governance review needs
  • Pilot requirements
  • Procurement questions
  • Recommendation for next step

References

  • Read references/ai-vendor-evaluation-checklist.md when comparing AI vendors, tools, platforms, or agent systems.