HomeEcosystem & integrations
Ecosystem

What plugs in, and what it connects

30 integrations across 5 categories. The four lines are written in different languages, but their infrastructure choices converge on one set of standard parts — which lowers the operational learning cost for our customers.

Linkage

How the four business lines connect to each other

This is our biggest difference from single-product companies: not four isolated systems, but one kernel with three applications on top.

UIAMUniclaw
Identity issuance

Agents register as MACHINE principals; credentials, scopes and tool permissions are issued by the identity kernel and can be revoked instantly

UIAMUniscrm
Employee identity

The internal account system connects to unified identity, converging login, roles and permissions in one place

UIAMUnilearning
remote mode

Switching the auth domain from local to remote completes the integration, with no changes to any other domain

UniclawUniscrm
Digital workers

Agents handle common enquiries in service scenarios, and past conversations become their knowledge source

UniclawUnilearning
AI grading

One LLM abstraction layer: Unilearning uses it to draft grading, and the teacher makes the final call

UniscrmUnilearning
Customer training

Private-domain customers can become learners, with orders and progress unified in one customer view

uiam-connector

Identity sources & social login

6 items

Domestic IM ecosystems and international social accounts run through one connector abstraction, so external identities can be mapped onto local identities and identity tags.

WeCom
Identity source

QR-code login and directory sync; the org structure becomes the identity source, and join/leave events drive account lifecycle.

Integration modeOAuth2 authorization code + directory callback
WeChat Open Platform
Identity source

WeChat login and user profile retrieval, unifying consumer membership across platforms.

Integration modeOAuth2 authorization code
Mini programs
Identity source

Mini-program session exchanged for a unified identity, so app, mini program and web share one membership.

Integration modecode2session + identity binding
WeCom OA
Identity source

Internal office integration, with employee records and org relationships as the authoritative source.

Integration modeIncremental directory sync
GitHub
Identity source

Social login for engineering collaboration, used by internal tools and external partners.

Integration modeOAuth2
Google
Identity source

International social login, sharing one connector admin console with GitHub.

Integration modeOAuth2 / OIDC
uniclaw-channel · uniscrm

Collaboration & reach channels

6 items

Wherever users are, the agent is there too. Channel adapters normalise each platform’s messages into one internal message structure, so business logic never perceives platform differences.

WeCom (SCRM)
Channel

Connected in SCRM pull mode to reach customers inside the WeCom ecosystem; Uniscrm also handles external contacts and conversation archiving.

Integration modepull + callback
Feishu
Channel

Push API for outbound messages plus event reception — suited to internal assistants and approval flows.

Integration modeEvent subscription + message push
DingTalk
Channel

Push API for outbound messages, sharing one channel adapter abstraction with Feishu.

Integration modeEvent subscription + message push
Web Chat
Channel

Real-time bidirectional WebSocket communication, embeddable in your own product as a conversation entry point.

Integration modePersistent WebSocket connection
Open API
Channel

Called directly by business systems to embed agent capability into existing product flows.

Integration modeREST / streaming SSE
Mini programs & H5
Reach

Short-link redirects, mini-program configuration and H5 pages form the private-domain distribution chain.

Integration modeShort link + mini-program configuration
uniclaw-llm · unilearning-ai

Models & intelligence

6 items

Model vendors are pluggable, with no lock-in to any single one. Pick the model per scenario: fast models for conversation, strong models for planning.

OpenAI-compatible protocol
Model

A unified abstraction in the llm module: any service speaking the OpenAI protocol can connect, including privately deployed ones.

Integration modeHTTP / streaming
Anthropic
Model

Optional for long-context and tool-calling scenarios, sharing one calling abstraction with other vendors.

Integration modeVendor adapter
Google
Model

Used for multimodal and retrieval-augmented scenarios, with available models configured per tenant.

Integration modeVendor adapter
GLM / DeepSeek
Model

Domestic models connect over the compatible protocol; the default candidates for AI grading.

Integration modeOpenAI-compatible
Local Ollama
Model

Local inference for fully intranet scenarios, so grading depends on no external service.

Integration modeOpenAI-compatible
Milvus
Vector store

The vector retrieval foundation for memory and knowledge, privately deployable, with vectors never leaving the intranet.

Integration modegRPC
Shared across product lines

Data & middleware

6 items

The four business lines speak different languages, but their infrastructure choices converge on the same set of standard parts, lowering the operational learning cost for customers.

MySQL 8
Database

The primary database for UIAM and Uniscrm, with migration files maintained per dialect.

Integration modeJDBC + Liquibase
PostgreSQL
Database

Supported homogeneously with MySQL; PostgreSQL is the preferred dialect for Unilearning.

Integration modeJDBC / gorm
Redis
Cache

Sessions, tokens and rate-limit counters, supporting the low-latency token validation requirement.

Integration modeRESP
RabbitMQ
Messaging

Async tasks and message delivery, absorbing peaks in WeCom sync and notifications.

Integration modeAMQP
Object storage
Storage

Assets, media and courseware resources; Unilearning supports real AssumeRole temporary credentials.

Integration modeS3-compatible / OSS STS
Docker
Delivery

Delivered as a single image or Compose with consistent environments — the standard private-deployment form.

Integration modeImage / Compose
integration · runtime

Gateways & runtimes

6 items

Authentication can move forward to the traffic entry point, and each product line picks the runtime shape that fits it best rather than being forced into uniformity.

Higress
Gateway

Token validation happens at the gateway, so business services only see authorized identities and authentication moves to the traffic entry point.

Integration modeWasm plugin / external authorization
MCP
Protocol

The standard integration protocol for AI agents: UIAM authorizes tools, Uniclaw calls them, NewTool exposes algorithms.

Integration modestreamable HTTP
Kotlin / Spring Boot
Runtime

The runtime for the identity kernel and the WeCom server side, using virtual threads for high-concurrency authentication.

Integration modeJVM
NestJS / TypeScript
Runtime

The agent platform runtime, with LangGraph driving the cognition-loop state machine.

Integration modeNode.js
Go
Runtime

The modular monolith and the capture kernel, delivered as a single binary — fast to start, few dependencies.

Integration modeStatic compilation
Rust / WASM
Runtime

The algorithm kernel compiles to native and wasm targets, so one implementation serves every platform.

Integration modeNative / wasm32
Product Lines

Integration capabilities by business line

Each product line’s detail page also carries its own list of integrations.

Get in touch

Hand the complexity of identity, agents and private domain to one governable kernel

Whether you are replacing an existing IAM, building an agent platform, or trying to make private-domain operations actually work — start with a 30-minute architecture call. We will first judge whether this is the kind of problem we are good at, and say so plainly if it is not.