HTTP Agent that keeps socket connections alive between keep-alive requests. Formerly part of mikeal/request, now a standalone module.

Playwright Tools for MCP

A lightweight registry of LLM model information, like name and context sizes, for building AI-powered apps.

A PAC file proxy `http.Agent` implementation for HTTP

SDK for building AI agents with Claude Code's capabilities. Programmatically interact with Claude to build autonomous agents that can understand codebases, edit files, and execute workflows.

Token efficient IDs for AI agents. UUID alternative for the agentic era

Zero-dependency OTLP span contract for agent traces: span shape, gen_ai attribute vocabulary, pure builders, and a conformance validator that reports what a trace can and cannot answer

Algorithmic Agent Harness (ADR-047) — a deterministic control plane that wraps any model/tool/skill as a replaceable worker: utility-scored decisions, four-gate safety, hash-chained receipts, circuit-breaker recovery, adversarial verification, and a bandi

Loads environment variables from .env file

React integration for LangGraph & LangChain

Subagent dispatch with live TUI streaming and cost tracking

SDK to facilitate Node integrations with Stark Bank

Custom metrics for AWS

Performance measurement

Native CLI binary for @anthropic-ai/claude-agent-sdk on linux-x64

Fresh-context iterative loops for Claude Code — autonomous task completion with independent verification

Open-source-safe CLI for persona simulation, observer review, and public-safe feedback drafts.

Detect whether a JavaScript CLI is running inside a coding agent.

Lightweight agent orchestration plugin for OpenCode - a slimmed-down fork of oh-my-opencode

Eval harness for Last Light workflows — drives the real production workflows against a mocked GitHub and grades deterministically (SWE-bench compatible).