> For the complete documentation index, see [llms.txt](https://help.cubelet.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://help.cubelet.ai/platform-and-integration/mcp-servers/manufacture-cubelet-manufacturing.md).

# manufacture — Cubelet Manufacturing

*The platform's primary MCP server: generic Cubelet content, concept navigation, cross-domain simulation, and the CMMC / ISACA / AI Governance / GRC domain suites at one endpoint.*

## Endpoint

`https://manufacture.cubelet.ai/mcp` — streamable HTTP. Every tool call requires authentication; see the Auth & access page for how to connect.

## What it serves

`manufacture` carries the tools that work the same way regardless of which domain you're querying — CMMC, ISACA Tri-Cert, AI Governance, NIST Privacy Framework, VulnOps, and GRC Engineering all go through the same tool names. Since August 2026 it **also** carries the CMMC, ISACA, AI Governance and GRC domain suites that were previously dedicated-worker-only. VulnOps tools are the one exception and stay on the VulnOps worker — see the MCP servers page.

### Cubelet content

Read and search access to the Library's Cubelets:

* `cubelet_get` — get a Cubelet by ID; optionally filter to a single face (WHAT, WHY, HOW, WHERE, WHEN, or APPLY).
* `cubelet_list` — list Cubelets with scores and metadata, optionally filtered to one domain slug.
* `cubelet_search` — search competencies and Cubelets in fulltext (default), semantic (vector), or ontology (taxonomy + vocabulary) mode, with optional thesaurus query expansion.
* `catalog_cubelets` — return the public Cubelet Catalog organized by AREA (CMMC L2, ISACA certs, AI Governance, VulnOps) → Domain → Practice. Only taxonomy-registered Cubelet IDs appear.

### Concept navigation

Knowledge-graph tools for walking the taxonomy a Cubelet belongs to:

* `concept_navigate` — navigate the knowledge graph for one concept. `resolve` returns its hierarchy breadcrumb, definition, and the quality-passed Cubelets that teach it; `associate` returns its related neighbour concepts within the same domain.
* `concept_graph` — return a concept's ego-network: its prerequisite chain, what it unlocks (dependents), and related concepts across domains — a pre-assembled context bundle for agent synthesis.
* `concept_translate` — find semantically equivalent or related concepts across compliance frameworks (for example, a CMMC control's nearest NIST AI RMF or EU AI Act equivalents) via vector similarity, returning source Cubelets for grounding.

### Domain reference, consulting and reporting

Added to this server in August 2026. Entitlement gates apply exactly as they do on the dedicated workers.

* **CMMC** — `cmmc_domain_get`, `cmmc_practice_get`, `cmmc_practice_list`, `cmmc_practice_search`, `cmmc_gap_analysis`, `cmmc_sprs_score`, `cmmc_poam_generate`, `cmmc_evidence_guide`, `cmmc_readiness_report`, `cmmc_coach_explain`, `cmmc_coach_recommend`, `cmmc_tier_capabilities`, `cmmc_tier_compare`.
* **AI Governance** — `ai_gov_framework_ref`, `ai_gov_traiga_ref`, `ai_gov_tools_ref`, `ai_gov_report`.
* **ISACA** — `gjf_get_priority_stack`, `gjf_get_lens`, `isaca_report_export`.
* **GRC** — `stix_coverage`, `thesaurus_negotiate`.

### Simulation tools (Case / Lab)

These back the practice experience learners work through in the Lab — distinct from the Cubelet content above, which is studied in the Library. Simulator output is never called a Cubelet. Most of the groups below run a Case: a judgment exercise against a fictional organization, scored against ground truth. The ISACA QAE tools are a separate, lighter-weight practice format and are not a Case — see the note on that bullet.

* **ISACA simulation** (`isaca_simulation_*`) — Case-format sessions: the learner interviews a fictional organization, requests evidence, and is scored on judgment, lensed to a chosen ISACA Tri-Cert certification (CISA, CISM, CRISC, or CDPSE): `isaca_simulation_start`, `isaca_simulation_interview`, `isaca_simulation_evidence`, `isaca_simulation_score`, `isaca_simulation_status`, plus `isaca_mirror_moment` for post-score feedback.
* **ISACA QAE** (`isaca_qae_*`) — adaptive multiple-choice practice questions for the ISACA Tri-Cert certifications: `isaca_qae_start`, `isaca_qae_answer`, `isaca_qae_reveal`, `isaca_qae_status`. This is a standalone practice format, not a Case.
* **AI Governance simulation** (`ai_gov_*`) — role-lensed judgment simulations against AI-governance scenarios: `ai_gov_scenarios`, `ai_gov_get_lens`, `ai_gov_start`, `ai_gov_message`, `ai_gov_evidence`, `ai_gov_score`, `ai_gov_mirror`, `ai_gov_status`, `ai_gov_list`, `ai_gov_abandon`.
* **CMMC simulation** (`cmmc_simulation_*`) — mock CMMC Level 2 assessments where the learner plays assessor or auditee against an AI-driven NPC organization: `cmmc_simulation_start`, `cmmc_simulation_interview`, `cmmc_simulation_evidence`, `cmmc_simulation_score`, `cmmc_simulation_status`.

The lists on this page are a guide, not a contract. Call `tools/list` against the endpoint for the authoritative surface.

## Quality-gated serving

Only Cubelets that have passed the quality gate — 42/60 aggregate minimum, 6+ points on every face, no face scoring 0 — are discoverable or retrievable through this server. `cubelet_list` and `catalog_cubelets` filter unpassed Cubelets out of their results before returning them; `cubelet_get` returns an error rather than content for a Cubelet that hasn't passed. There is no path on this server to retrieve unpublished or failed Cubelet content.

## Authentication

Every tool on this server requires an authenticated request. See the Auth & access page for setup.