MCP

The MCP is the agent-facing surface for helping an agent author a policy: the guardrail a user’s intent is built on. It grounds that policy in real categories and evidence attributes from the product taxonomy, then validates it with the orchestrator.

Connecting

The MCP is served over streamable HTTP at the /mcp path of your customer-specific delta endpoint:

https://<your-host>/mcp

It is stateless. Point any MCP client at that URL.

Tools

• get_policy_language_reference (no args) — returns the full reference for the policy language.
• search_taxonomy { query? } — browses or searches product categories in the taxonomy, returning the categories and attributes that a policy can refer to in its evidence. Omit query to list top-level categories; pass a category exactly as returned by a previous call to see sub-categories or, for a specific product, its attributes; pass any other text to run a semantic product search.
• validate_policy { source } — validates a policy: compiles it and returns its ID on success. Returns a compile error if the policy is invalid. Run it on a finished draft and after every edit.

The policy language reference is also exposed as an MCP resource at doc://policy-language-reference. Reading that resource returns one text/markdown body with the same bytes as get_policy_language_reference, so clients that surface resources can attach it directly without a tool call.

Authoring loop

1. Read the policy language reference with get_policy_language_reference.
2. For each product type the policy covers, call search_taxonomy with a description to find its taxonomy category and evidence attributes; if a search finds nothing, browse instead — no query lists the root categories, then follow category paths down.
3. Draft the policy.
4. Call validate_policy on the draft; fix and repeat until it succeeds. On success the policy compiles and you get back its policy ID.
5. The user then signs an intent referencing that policy ID and submits it — see the Orchestrator API.