Produced by k8s-platform/tools/generate-mcp-tools.py: deck file openapi2mcp
does the conversion, and the wrapper re-injects the five things Kong's
generator cannot emit — per-tool ACLs, the MCP behavioural annotations, the
response schemas the plugin uses for structuredOutput, the real upstream host,
and JSON-only request bodies. Each is a documented gap; see
k8s-platform/MCP-TOOL-GENERATION.md.
tools-policy.yaml carries the ACLs and doubles as the allow-list: an operation
becomes a tool only if the policy names it. Kong would otherwise expose every
operation in the spec, DELETE included, and can express no authorization at all
from a bearer-token spec — silently, exit 0, no warning.
Same tools as before, same ACLs. What changed: query parameters are now
declared and callable, each tool carries its response schema so calls return
structured content instead of raw JSON, and request bodies are no longer
tripled across three identical content types.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
16 routes: Office (single key, full CRUD + employees sub-resource)
and Employee (single key, full CRUD + reports + customers
sub-resources). MCP layer built with the single-route pattern from
the start.
Only domain where customer role has zero access — verified against
ReferenceDataPermission.check(): read_only/product_manager/support
get reads, writes are admin-only, customer isn't in the allow-list
at all. All 9 MCP tool ACLs reflect this exactly.
Keycloak (local): workforce-api-audience + mcp-workforce-audience
client scopes, resource clients, broker/librechat attachments.
Deploy key and broker secret reused/mirrored, same pattern as every
domain since payments-api.
Stéphane Tailland