Search

OpenClaw plugin: turn vague requests into execution-ready briefs. Adds 3 agent tools — ambiguity scan, plan stress test, and brief generator. No API key. No external service.

Self-hosted autonomous AI agent for social media. Per-platform voice memory, 11-tool agentic loop with error recovery, vLLM local inference. Publishes to X, LinkedIn, Facebook, Instagram, Threads from one prompt.

Rust multi-agent LLM framework — OpenClaw alternative. WhatsApp + Telegram + Gmail + browser agents on a single binary. NATS broker, MCP client+server, Claude/Anthropic OAuth, MiniMax, durable TaskFlow workflows.

Self-tuning multi-agent AI trading system. 8-source signal fusion (Polymarket + Kalshi + 10 ML models incl. Kronos foundation model), Bull/Bear/Judge debate on Claude Opus 4.7, Portfolio Manager gate.

Community-maintained directory of EU-reachable job sources for agentic job search — machine-readable, with coverage, access type, rate limits, and real-world crawl notes.

Agent harness for hot-reloading function evolution in Rust.

CloudSRE v2 is a 16-service asynchronous RL environment where a 1.5B agent learns to resolve cascading microservice outages against a 72B Adversarial Designer.

OmniTrade: LLM-driven crypto futures arena — 11 competing strategies on Agno + AgentOS, atomic three-way state, real-time SSE dashboard.

Fleet court — dispute resolution and arbitration for multi-agent coordination

Model-agnostic Python SDK for building production-grade agents against any LLM provider — Anthropic, OpenAI, or local Ollama.

Rust runtime for Python AI apps. Drop-in for openai/anthropic SDKs with native SSE streaming, an agent loop with concurrent tool dispatch, and Logfire-compatible OTel emission.

Every AI tool forgets you after every session. This repo doesn't. Ever. Zero install, zero config. Just fork and your AI remembers everything, everywhere, forever.

Framework-agnostic confidence-gated escalation middleware for LLM agents: multi-signal scoring (logprob, verbalized, tool risk), threshold policies, and escalation handlers for LangChain, CrewAI, AutoGen, and Google ADK.

Chain-Receipt SDK — emit, verify, replay, and chain Receipts (v1.0.0) from LangChain runnables, agent loops, or any Python callable. Methodology dependency for the Calibrated Chain-Determinism paper.

Comprehensive best-practices knowledge base — installable as an AI agent skill with auto-injection of defaults into Copilot CLI and Claude Code

Unofficial Octoparse CLI for AI agents and automation workflows

Shared browser interaction schema registry for AI agents. Reduces LLM token usage by 80-100% on known sites.

Vaultak runtime security callback handler for LangChain agents

Pi extension that gives non-vision models the ability to analyze images via a vision-capable subagent

GAPA — Generalized Action and Prompt Adaptation. A cross-IDE self-learning framework for AI Agents. Supports Kiro, Cursor, Claude Code, VS Code (Copilot), Windsurf, and TRAE.