Sandboxed TypeScript runtime for AI coding agents to query AWS — full SDK access with in-sandbox aggregation, credentials flow via IMDS and never enter the VM. Ships as a Claude Code plugin (MCP) and a standalone CLI, backed by Shuru microVMs or Docker.
Self-hosted operations backend in a single Go binary — short links, email and DNS ship as plugins, and every capability is drivable by AI agents over MCP.
termcp is an MCP server written in Go that exposes interactive programs to AI Agents as persistent SSH sessions, letting Agents continuously manage and drive them. On top of that, termcp ships a dedicated session management UI that gives you full visibility into the Agent's behavior.
AI-native developer observability MCP (Model Context Protocol) server that gives AI agents full-stack visibility into your development environment — browser, terminal, and processes — all in one unified interface.
Makes AI stateful across sessions. Workflow: chat → idea → design → reqs → plan → do_step(s) → done, with fresh scoped context, staleness detection, and human approval at every phase. Roadmap & history of shipped features.
Run, watch, and meter Claude Code, Codex CLI, Gemini CLI, OpenCode, and Kiro from one CLI. Normalized events, cache-aware token usage, cost accounting verified against each agent's own records, budget caps, OpenTelemetry gen_ai spans, Langfuse traces, and a live web dashboard + MCP server.
End-to-end tests without test code. Write one sentence, an agent drives a real browser on every pull request and tells you what broke. A passing run is recorded and replays with no model call. Commercial, hosted at smolanalytics.com.
One MCP server for all web data — search, scrape, crawl, proxy, and AI research in a single npx install. Works with Claude, Cursor, and any MCP client.
MCP server + CLI that puts a real human in your coding agent's loop. It publishes work mid-task, a reviewer pins what's wrong and returns a verdict, and the agent iterates until approved.
A thin MCP client for humans and agents — run any Model Context Protocol server, call its tools, read its stderr, and reconnect after each edit. REPL, headless CLI, or an MCP server your agent drives. No client config edits, no restarts.