Open spec for how services communicate operational limits to humans and agents: limits discovery, structured refusals, conformance levels.
Access the llms.txt files for this website
Graceful Boundaries is an open specification (CC-BY-4.0) for how HTTP services communicate operational limits — rate limits, quotas, auth, payload size — to humans and autonomous agents. It extends RFC 9457 problem-details with a structured refusal vocabulary, defines a limits-discovery endpoint, and specifies four conformance levels (L0–L4).
npx graceful-boundaries check <url>The llms.txt at gracefulboundaries.dev points to the spec, the RFC 9457 profile, the conformance checker, and worked examples — the minimum context an agent needs to either implement the spec or interpret a conformant service's responses.
Explore tools created to help you work with llms.txt
Discover similar websites implementing llms.txt
Production-readiness gate for AI-built apps. VibeRaven gives coding agents a launch checklist, prioritized gaps, agent tasklists, and deploy verification before auth, billing, database, RLS, env vars, webhooks, monitoring, or tests.
Open-protocol directory of AI-native websites. Verifies llms.txt and agents.json compliance, issues trust badges, and makes verified sites queryable by AI agents and IDE assistants.
Micra.js is a ~5 KB reactive UI library for server-rendered pages: reactive `state` via a shallow `Proxy`, declarative `data-*` and a cross-component event bus.