# Catalogue

> Public catalogue components.

JSON: https://leyline.ahensya.com/api/v1/components

- [TinyFish](https://leyline.ahensya.com/components/tinyfish.md): Search, fetch, and browser automation for agents. Free Search and Fetch; metered browser runs. Leyline uses it as intake evidence crawl, not as a locked vendor.
- [GitHub](https://leyline.ahensya.com/components/github.md): Repo host whose stars, commits, and issues feed Leyline GitHub health — a Score factor, not the whole reputation story.
- [Playwright](https://leyline.ahensya.com/components/playwright.md): Browser automation for tests and agents. Use the browser slice without taking the whole test-runner product box.
- [OpenRouter](https://leyline.ahensya.com/components/openrouter.md): Unified model and inference router. Listed honestly as an inference gateway — inspiration-adjacent, not Leyline's core.
- [Browserbase](https://leyline.ahensya.com/components/browserbase.md): Hosted browsers for agents. Browser slice without running Playwright locally.
- [LangSmith](https://leyline.ahensya.com/components/langsmith.md): Tracing and evals for LLM apps. Observe and evaluate slices that feed Score later.
- [agent-desktop](https://leyline.ahensya.com/components/agent-desktop.md): Gives agents structured desktop accessibility reads and actions through a CLI, C library and MCP server.
- [MCP / AI Gateway](https://leyline.ahensya.com/components/mcp-ai-gateway.md): Gateway layer that routes model calls, discovers MCP tools, and applies identity, policy, and spend controls before invocations.
- [Toloka](https://leyline.ahensya.com/components/toloka.md): Human-in-the-loop training data, RL environments, and evaluation for AI agents and LLMs.
- [cloro](https://leyline.ahensya.com/components/cloro.md): Returns structured search and AI-response results, including available source citations.
- [telemetry.dev](https://leyline.ahensya.com/components/telemetry-dev.md): Hosted observability for AI/LLM apps. Product line: every model call, tool step, and retrieval in one OpenTelemetry trace, with tokens, cost, latency, and errors. Ingest via TypeScript or Python SDKs, or any OTLP exporter. Cost is computed server-side against 4700+ models.