Search

Evidence-driven AI application diagnosis and regression gating for RAG, agents, and inference.

Worktree multiplexer for Claude Code — one git worktree + one agent per chat session

Automate Blender 3D modeling with AI via MCP — 93 tools for modeling, sculpting, architecture/MEP, materials, and 3D-print prep. Includes Blender Studio: a visual record/replay session editor with a multi-provider AI assistant (Claude, Gemini, OpenRouter) — drive it from n8n, Claude Code, or standalone.

An AI agent you build, not just prompt. Local-first Android agent with on-device inference (LiteRT-LM), pipelines you assemble yourself, MCP tool calling, and your confirmation before any risky action.

Spec-first, embeddable agent harness runtime - a control plane for loop-engineered AI agents. Governed tool calls through one gateway, one normalized event stream, durable sessions.

Koder — aiXplain's agentic coding environment. Your terminal, your tools, your models.

Stateful Node.js sessions for Codex, Claude, Grok, Bedrock, and Responses-compatible model providers.

A lightweight runtime that unifies 110+ AI providers behind a single, OpenAI-compatible API.

pi extension: fast in-process subagents with a dependency-graph scheduler (needs edges gate tasks and carry upstream output into dependent prompts), plus background runs, intercom and agent-to-agent mailbox. Leader defines agents inline.

Multi-agent collaboration relay for DeepSeek Harness — send and receive messages between local agents (dsh, Codex, Claude, Hermes, OpenClaw, …) through a small loopback-first broker with HMAC auth. Ships the broker, the dsh plugin, a CLI client and a Pyth

npm install shim for Jacky CLI (self-improving AI agent) — installs the Python package into a managed venv under ~/.jacky and exposes the `jacky` command.

Run Corsac's eval agent against your AI product: generate scenario-based evals and grade them with an LLM judge.

Run agent-browser in a Vercel Sandbox created from a browser-ready container image

Understand any GitHub repository with AI - analysis plus evidence-based agent chat.

Local identity for an agent: wallet, secrets, and optional email

A local-first personal assistant: pluggable model providers, three-layer memory, tracing and evals.

Execution cache layer for AI agents — cut LLM and tool call costs by reusing previous results.

AI model: ermiaazarkhalili/Agents-A1-4B-SFT-Fable5-Glint-GGUF

AI model: mradermacher/AgentMercury-Qwen3.5-4B-GGUF

Local-first AI Agent framework in Java. ReAct loop, MCP tools, layered memory, multi-agent orchestration — run via Shell, Web or embed it.