Search

Agentic Control Plane for DeepSeek Harness — check every tool call against your policies before it runs, and keep a durable record of what was allowed and why.

Catch the step where an AI agent contradicts a decision it already made. Adapters and a CLI for the Fathom committed-state read over LangGraph, CrewAI, Letta, OpenInference, DBOS, and coding-agent traces.

Open-source, self-hosted AI debugging and observability engine for LLMs, RAG, and autonomous agents.

Production-grade personal AI assistant — full agent loop, multi-LLM routing (Claude/GPT/DeepSeek/Kimi), 50+ MCP tools, tiered safety model, multi-entry (Web/TUI/Voice).

AI Native Security Tool for Wallet && Password management.

🌍 Agentic geopolitical risk analyzer — discovers second-order asset exposures through historical evidence, not LLM hallucination. Agent proposes, deterministic layer verifies. LangGraph + ChromaDB + FastAPI.

A unified evaluation framework for agents across LLMs and agent versions

Official Python SDK for QAISS — Non-blocking telemetry, observability, and passive auditing for AI agents and LLMs.

LiteLLM provider backend for purrcept_core.

Automatic LangChain Agent observability for AgentBase AgentOps

Discover and manage Claude Code, Codex, Gemini CLI, and other Agent Skills inside DeepSeek Harness.

Install curated AI agent skills for what your project actually uses

Technical specifications and legal compliance suite for AI agents

Fail-closed network recovery, recommended answers, and safe unattended continuation for DeepSeek Harness.

A voice-first personal agent for developers: one continuous, local-first collaborator in your terminal.

CLI that automates the initial setup of Coderium Agent Framework (CAF) in any repo

Your AI software company. A CLI that spawns AI agents (Planner, Designer, Developer, Tester, Reviewer, BugFixer) to build software from a single idea. Model-agnostic: works with GPT, Codex, GLM, Gemini, DeepSeek, Qwen, Claude and local models.

Agent-native RCS messaging with automatic SMS fallback. Open a conversation, get typed replies, and let the channel flip handle itself.

Self-hosted AI chat dashboard for Hermes Agent — multi-model web UI with multi-platform integration

从设备信号判断:有没有发生值得留意的事、值不值得叫醒一次 agent、以及该怎么把此刻的状况讲给它听。不采集数据、不选数据库、不调模型 —— 存储由宿主实现 StoragePort，编排在包内。