ClearPoint Logic is built in layers. Nexus helps teams adopt proven skills. Studio turns new ideas into real agents. Meridian governs the workforce record. Under all three, Anchor signs the proof and Agent Core runs the work.
The same architecture diagram our engineers walk customers through on day one.
Anchor is the layer everything in the platform is signed against. It ties people, agents, models, prompts, tools, datasets, and runs together into a single signed record that is auditable end to end and exportable to your evidence system.
The standard itself is open through Open Anchor, so anyone can read it, validate against it, and verify a claim independently. Anchor Pro, the trust fabric inside Meridian, is how ClearPoint implements and enforces it.
Agent Core runs your agents in whichever framework they were written in. ADK Go, LangChain, Anthropic Agent SDK, and Microsoft Agent Framework share the same identity, lifecycle, approvals, evidence, and memory.
CPL Agents are the actual implementations: code, prompts, tool bindings, and AI BOM. They’re published as signed packages, certified against the Anchor standard, and loaded at runtime into Agent Core or a customer-deployed Meridian instance.
Multi-framework: an agent can be authored in ADK Go, LangChain, Anthropic SDK, or Microsoft Agent Framework. Same publishing pipeline, same passport, same evidence. The framework is an implementation detail; the contract with the platform is not.
{
"name": "invoice-triage",
"version": "1.4.0",
"framework": "anthropic-sdk",
"passport": "cpl:passport:8f2a91c4",
"bom": {
"models": ["claude-3-5-sonnet@20251022"],
"tools": ["sap.lookup", "erp.write", "pii.redact"],
"datasets": ["finops.invoices.v3"],
"prompts": ["sha256:a4f2…"]
},
"scopes": ["read:invoices", "write:erp"],
"policy": "envelope://acme/finance/tier-1",
"signature": "anchor:sig:rsa256:c7d1…"
}Agents arrive from three places: your vendors’ clouds, your own builds, and ClearPoint itself. Vendor governance tools only ever see their own ecosystem. Meridian sits above all three and gives every agent one passport, one policy envelope, and one evidence layer, so the workforce record is complete no matter where an agent runs.
Agents that run inside a vendor’s own cloud and runtime. Meridian governs them through each vendor’s native control surface.
Agents you build and run in your own environment. Bring your own agent; it gets the same passport, policy envelope, and evidence trail.
Agents ClearPoint builds and runs for you, governed under the same single record as everything else.
On the roadmap: AWS · Salesforce · ServiceNow · SAP. Meridian also routes to leading models like Anthropic, OpenAI, and Gemini. Those are governed in the request path, never depicted as standalone estates.
ClearPoint builds on Google Cloud and partners with Anthropic. That’s how we are built, not a constraint on you. Meridian itself stays neutral: it governs Microsoft, Google, and the custom agents your teams bring or build under one record. We run our own AI workforce on ClearPoint too, as operating discipline, not as a constraint on your stack.
The platform itself is GCP-native. Agent Core runs on the Google Cloud control plane; ADK Go is a first-class runtime path; Gemini is integrated as a reasoning option alongside Anthropic Claude.
Claude is the primary reasoning model in our agent stack. We integrate with the Managed Agent Platform and align with the Anthropic Partner Network for distribution and reference patterns.
Read the strategy brief for the architectural philosophy, or register interest for a technical walkthrough of the platform layers.