architect.validate
Validate code, workflows, or architecture against AI Design Blueprint principles. Get principle coverage, findings, and actionable recommendations. Enterprise-safe with transient processing.
Instructions
Pro/Teams — evaluate code, a workflow, or an architecture description against the Blueprint doctrine. Returns principle coverage, findings, and example recommendations. ENTERPRISE-SAFE: payloads are processed transiently in memory by the underlying LLM provider (OpenAI API, no-training-on-API-data) and dropped. We never train models on user code, payloads, or architecture diagrams. Pass private_session=true to force the MCP server to bypass all database logging for this call — enforced in code, not just in policy. UK/EU data residency. DPAs available for Teams. Auth: Bearer .
Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| implementation_context | Yes | Code snippet, workflow description, or architecture summary to evaluate against Blueprint principles. | |
| focus_area | No | Narrow the evaluation to a specific principle cluster or slug (e.g. 'delegation-and-scope'). | |
| task | No | What the agent or workflow is trying to accomplish. Adds evaluation context. | |
| language | No | Programming language of the code being evaluated (e.g. 'python', 'typescript'). | |
| repository | No | Repository name or path for additional context. | |
| files | No | List of file paths relevant to the implementation context. | |
| goals | No | Specific safety or quality goals to evaluate against (e.g. 'prevent irreversible actions', 'explicit approvals'). | |
| example_limit | No | Maximum number of curated examples to include in recommendations. | |
| private_session | No | Set to true to disable all logging for this validation call. |