# Catalogue

> Public catalogue components.

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

- [Fix Cold Start Latency in AI Sandboxes: Techniques](https://staging.leyline.ahensya.com/components/fix-cold-start-latency-in-ai-sandboxes-techniques.md): # How to fix cold start latency in AI sandboxes: techniques and tradeoffs
- [Magika](https://staging.leyline.ahensya.com/components/magika.md): Magika (google/magika) is Google’s AI-powered file content-type detector: it classifies files from byte content (not just extensions / magic bytes), locally, in milliseconds on CPU. First-party README: ~100M-sample training, 200+ content types, ~99% avg precision/recall on their test set; used at scale for routing Gmail/Drive/Safe Browsing samples to scanners; also integrated with VirusTotal and abuse.ch (README claims).
- [REA](https://staging.leyline.ahensya.com/components/rea.md): README tagline: "One MCP for reverse engineering across binaries, applications, and runtime behavior." The agent calls REA through MCP to inspect a local target and trace relevant code; REA returns findings with their evidence, and the agent uses them to explain the behavior or write and test an implementation.
- [AI Agent Sandbox Comparison 2026: Seven Platforms](https://staging.leyline.ahensya.com/components/ai-agent-sandbox-comparison-2026-seven-platforms.md): # The Ultimate AI Agent Sandboxes Comparison 2026
- [Best code execution sandbox for AI agents in 2026](https://staging.leyline.ahensya.com/components/best-code-execution-sandbox-for-ai-agents-in-2026.md): Published 17th January 2026
- [Onyx](https://staging.leyline.ahensya.com/components/onyx.md): Open-source enterprise search and AI assistant that connects to internal knowledge sources, indexes them with source permissions, and supplies humans and agents with answers and context. It is described as self-hosted in the customer's cloud, VPC, Kubernetes, or air-gapped enviro
- [RTK](https://staging.leyline.ahensya.com/components/rtk.md): A Rust-based CLI proxy that filters and compresses command output before it reaches an LLM or coding agent. The repo positions RTK as a low-overhead way to reduce token consumption on common development commands like git status, git diff, test runners, grep/search, logs, and cloud/devops tooling.