Search

Pure Python client (no LangChain) for Andreax's pay-per-call x402 AI tools (USDC on Base, EIP-3009 autopay).

Merge Gateway CLI — route your coding agents (Claude Code, Codex) through Merge Gateway

Local multi-provider LLM proxy, model router and control plane for Claude Code, Codex, OpenCode, Gemini CLI and other AI coding agents: 57 providers, fallback routing, credential health, a native web-search tool proxy, and analytics with cost provenance

Wires project Brainspaces and skills into coding agents.

Self-improving agents through continual learning: in-process agentic RL that turns your own agent's work into training data. Observed tool outcomes, outbound-only, one runtime dependency.

Hugging Face Space: chatbot_three

Run prompts and workflows approved in AirPrompter on your own systems: signed pull-only updates, an encrypted restart-safe slot store, offline operation, content-free telemetry.

Smell-check your prose. Installable writing rule set for AI agents that sprays the performative, faux-profound register off prose. Core bans, smell audit, claims review, genre overlays. Prose, not code smells.

Detect if code is running in an AI agent or automated development environment

opencode-style custom agents for dsh: define agents as markdown files with their own system prompt and model, then call them from chat with an @mention — each mention dispatches a subagent

An uncensored, local-first CLI coding agent — streams live reasoning, edits files, runs commands. Provider-agnostic, runs on any OpenAI-compatible uncensored LLM.

The run log as a browser reads it: a tenant-GraphQL RunLogStore over a module's agent_run/agent_event pair, plus the follower loop every run surface uses

The agent module's conversation as a library: thread load/create, an inbox over new messages, transcript append, ToolPart state machine and agent_task rows — over an injected GraphQL client.

Service-style agent kernel for runs, sessions, tools, suspension, and recovery

Python SDK for Golem extensions

Agent 能力评估系统 — 基于 Agent-Driven 架构的评测框架

Provider-agnostic reference commerce agent — a fork of anthropics/commerce-agents that runs on any OpenAI-compatible chat-completions endpoint (OpenRouter, Chutes, LiteLLM, vLLM, self-hosted).

ZhiHu CLI PRO — Cross-question deep research for Zhihu & AI Agents. Search, capture, verify, analyze large corpora, preserve evidence lineage, and synthesize across questions.

The host side of a headless agent run: a GateHost that asks for approval through the conversation thread, persona → model/prompt/tools selection, and an AgentEvent → transcript/task writer.

Record the OpenAI Agents SDK's own run structure — agents, handoffs, guardrails — into an OrcaReplay trace.