ClaraVerse vs mateclaw
Self-hosted private AI workspace: chat, multi-agent crews with human review, a visual workflow builder and Telegram integration in one app — local models via Ollama/llama.cpp or your own keys. — versus — Self-hosted Java agent platform: ReAct plus plan-and-execute 'digital employees' with an LLM Wiki knowledge layer, skills/MCP/ACP tools, approval-gated actions and eight IM channels in one JAR.
claraverse keeps everything local in a desktop workspace with Ollama/llama.cpp; MateClaw is server-first with multi-user governance and cloud-vendor failover — privacy-maximal vs. organization-ready.
| ClaraVerse | mateclaw | |
|---|---|---|
| Stars | 3.9k | 1.0k |
| Forks | 439 | 318 |
| Language | TypeScript | Java |
| License | NOASSERTION | Apache-2.0 |
| Last activity | 21 days ago | today |
| Topics | local, agents | agents, orchestration, memory |
| Curated connections | 3 | 5 |
ClaraVerse — the curator's take
The integration is the pitch: where a homelab stitches chat + n8n + agent runner, ClaraVerse is one AGPL app with crews, memory and workflows designed together — and it auto-detects Ollama/LM Studio with zero config. When NOT: one app means one surface to trust and its pace to accept; GitHub metadata says NOASSERTION while the README badges AGPL-3.0 — check LICENSE in the release you deploy; power users who want best-of-breed parts will feel boxed in.
mateclaw — the curator's take
Pick it when the buyer is an IT department rather than a hacker: multi-user workspaces, RBAC and approval gates on sensitive tool calls, audit trail, Spring Actuator health, per-channel error isolation, and provider failover across DashScope/OpenAI/Anthropic/Gemini/DeepSeek/Ollama when a vendor 401s mid-request. Underneath it's a real harness — StateGraph runtime, Team Runs with a durable runId, task DAGs, human approval — not a RAG call in a wrapper. The cost is weight and gravity: Java 21 + Spring Boot 3.5 + Vue, an enormous and fast-moving feature surface (workflows, triggers, Wiki transformations, Dream consolidation, skill mining), and a China-first channel set (DingTalk, Feishu, WeChat Work, QQ) whose value drops sharply outside that ecosystem. For a personal agent on a $5 VPS this is orders of magnitude too much machinery.