Search

AI model: fesalfayed/gpt-oss-20b-hermes_agent-tool-finetune_4bit

AI model: latte-agent/qwen3-4b-latte-v5

Async context manager middleware for LLM agents — solves Lost in the Middle via local Ollama compression.

Hard tool/MCP rerouter for opencode that prevents flooding and wasted context from searches, docs, web, GitHub examples, and verbose MCP calls.

⚡ Self-hosted AI agent framework — sandboxed plug-and-play WASM agents, a live code security radar (Threat Radar), Telegram-native bots & crypto/finance tooling. Single Go binary. Pair with Flowork Router for the full LLM gateway.

Agent-Native PPTX Engineering Library — Built for AI agent pipelines, not template filling. 22 unified layouts | 4 brand palettes | WCAG-AA color safety | Native CJK font support

Run any AI agent — Claude Code, GAP, LangGraph DeepAgents — in a hermetic sandbox. Reference implementation of the Harness Protocol.

Experimental autonomous work operating system focused on decision quality, human-AI collaboration, and evidence-driven autonomy.

Yim — A complete AI Agent project. This is a monolithic project containing Python agent framework, Rust native core, and Yim Desktop — not individual packages for separate installation. Backend-agnostic, multi-provider, with streaming tool calls, safety engine, persistent memory, multi-agent orchestration, and a rich desktop UI.

AI-native coding orchestration platform: unified multi-model agent runtime with stateful sessions, tool governance, and traceable delivery.

Runtime safety layer for LLM agents. Detects token spirals, kills doomed tasks early, tells you why.

Local-first security recon that briefs your AI coding agent: facts + tailored probe scripts, code-in / artifacts-out. No LLM, no server, no running app.

Thin, backend-neutral agent-harness middlewares for LangChain / deepagents: loop guard, binary-read guard, tool-result budget.

Python SDK for PACT-AX — multi-agent collaboration primitives

PreCompact ledger and per-component context profiler for Claude Code

PACT Agent Collaboration Layer - Multi-Agent Coordination Protocol

微秘书 (WMS) CLI — AI-first WeChat assistant tooling. Query contacts/rooms and send messages via the WMS open API.

Break your AI agent before your customers do. Stress-testing engine for AI agents.

Break your AI agent before your customers do. Stress-testing engine for AI agents.

Nodmix Enterprise Platform