--- title: "Reference Links" description: "Canonical standards, docs, specs, and source links for agent-readable software surfaces." url: "https://agentsurface.dev/docs/reference-links" lastVerified: 2026-09-24 lastModified: 2026-09-25T05:46:01.000Z --- This is the single source index for standards, specs, and official docs that Agent Surface tracks. Use it when you want the primary source rather than the curated tooling view. ## Agent-Readable Web [#agent-readable-web] | Reference | What It Covers | Use It For | | ------------------------------------------------------------------------------------------------------------ | ---------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------- | | [Cloudflare Agent Readiness](https://blog.cloudflare.com/agent-readiness/) | Discoverability, content accessibility, bot access control, and capabilities | Understanding what agent-readiness scanners currently check. | | [Agent Readiness](https://isitagentready.com/) | Public site scanner for agent readiness | Spot-checking whether a URL exposes agent-readable signals. | | [Is Agentic](https://is-agentic.com/) | Vercel public-site readiness reports powered by Ora | Inspecting evidence for public discovery, access, comprehension, and use. | | [Is Agentic developer docs](https://is-agentic.com/docs) | Read-only report API, CLI, remote MCP server, and agent skill | Retrieving and integrating completed public-site reports. | | [Lighthouse Agentic Browsing scoring](https://developer.chrome.com/docs/lighthouse/agentic-browsing/scoring) | Local, experimental Lighthouse category for WebMCP, agent accessibility, and discovery signals | Running an offline agent-readiness check, including on pre-release or authenticated environments. | | [AgentReady](https://agentready.org) | MIT-licensed, vendor-neutral spec of stable-ID requirements for agent find/read/act | Grounding requirements against a named identifier before trusting a scanner's opinion of it. | | [Shopify Product-Page Audit](https://www.shopify.com/agentic-readiness) | Commerce-specific structured-data check for a single product page | Spot-checking a storefront product page ahead of Agentic Storefronts. | | [WebMCP specification](https://webmachinelearning.github.io/webmcp/) | Page-scoped JavaScript tools through `document.modelContext` | Implementing the current browser-native tools proposal. | | [webmcp.dev](https://webmcp.dev/) | Earlier `@jason.today/webmcp` JavaScript library | Maintaining or evaluating that distinct legacy implementation. | | [Markdown for Agents](https://developers.cloudflare.com/fundamentals/reference/markdown-for-agents/) | `Accept: text/markdown`, `x-markdown-tokens`, `Content-Signal`, and edge HTML-to-Markdown conversion | Serving readable pages to agents without wasting context on HTML. | | [llms.txt](https://llmstxt.org/) | LLM-readable site indexes | Publishing a concise map of docs, APIs, and key resources. | | [Content Signals](https://contentsignals.org/) | Search, AI input, and AI training policy signals | Declaring how accessed content may be used. | | [Web Bot Auth](https://datatracker.ietf.org/doc/draft-meunier-web-bot-auth-architecture/) | Signed HTTP requests for bot identity | Letting sites distinguish accountable agents from anonymous automation. | | [API Catalog RFC 9727](https://www.rfc-editor.org/rfc/rfc9727.html) | `/.well-known/api-catalog` | Advertising API specs, docs, status pages, and related API metadata. | ## Agent Skills And Context [#agent-skills-and-context] | Reference | What It Covers | Use It For | | ------------------------------------------------------------------------------------------------- | --------------------------------------------------------------------------------- | --------------------------------------------------------------------------- | | [Agent Skills](https://agentskills.io/home) | Open format for reusable folders of instructions, scripts, references, and assets | Packaging repeatable agent capabilities without bloating the base prompt. | | [Agent Skills Specification](https://agentskills.io/specification) | `SKILL.md` frontmatter, directories, validation, and progressive disclosure | Authoring skills that portable agents can discover and load. | | [Adding Skills Support](https://agentskills.io/client-implementation/adding-skills-support) | Skill discovery, catalogs, activation, and context management | Implementing a skills-compatible agent runtime or client. | | [Agent Skills Reference Library](https://github.com/agentskills/agentskills/tree/main/skills-ref) | Validation and prompt XML generation tooling | Validating local or published skills. | | [AGENTS.md](https://agents.md/) | Repository instructions for coding agents | Giving agents repo-level commands, boundaries, conventions, and navigation. | ## MCP [#mcp] | Reference | What It Covers | Use It For | | ---------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------ | ---------------------------------------------------------- | | [MCP Introduction](https://modelcontextprotocol.io/docs/getting-started/intro) | MCP as the standard connector between AI apps and external systems | Grounding protocol decisions. | | [MCP Architecture](https://modelcontextprotocol.io/docs/concepts/architecture) | Clients, servers, hosts, and capability flow | Designing where MCP belongs in an agent system. | | [Build an MCP Server](https://modelcontextprotocol.io/docs/develop/build-server) | Server implementation basics | Exposing tools, resources, and prompts to agents. | | [Connect to Remote MCP Servers](https://modelcontextprotocol.io/docs/develop/connect-remote-servers) | Remote server connection model | Publishing MCP over network transport. | | [MCP Inspector](https://modelcontextprotocol.io/docs/tools/inspector) | Interactive server inspection and debugging | Testing tool definitions, schemas, prompts, and resources. | | [MCP Registry](https://registry.modelcontextprotocol.io/) | Public MCP server discovery | Finding and distributing MCP servers. | ## Auth And Identity [#auth-and-identity] | Reference | What It Covers | Use It For | | -------------------------------------------------------------------------------------------------------------- | --------------------------------------------------------------------------------- | -------------------------------------------------------------------------------------------- | | [OAuth Protected Resource Metadata RFC 9728](https://www.rfc-editor.org/rfc/rfc9728.html) | `/.well-known/oauth-protected-resource` | Letting agents discover authorization servers and resource metadata. | | [OAuth 2.1 Draft](https://datatracker.ietf.org/doc/draft-ietf-oauth-v2-1/) | Modern OAuth profile | Designing delegated machine-to-machine auth. | | [DPoP RFC 9449](https://www.rfc-editor.org/rfc/rfc9449.html) | Proof-of-possession access tokens | Reducing bearer-token replay risk for sensitive tools. | | [WorkOS auth.md](https://workos.com/auth-md) | Markdown-based agentic registration discovery | Letting agents register or claim user-bound credentials without an interactive sign-up form. | | [workos/auth.md Reference Implementation](https://github.com/workos/auth.md) | Example `auth.md`, app-side registration endpoints, and provider-side ID-JAG flow | Implementing provider-attested and OTP-backed agent registration. | | [ID-JAG Internet-Draft](https://datatracker.ietf.org/doc/html/draft-ietf-oauth-identity-assertion-authz-grant) | Identity Assertion JWT Authorization Grant | Evaluating provider-attested agent identity assertions. | ## Agent Runtimes And Tooling Docs [#agent-runtimes-and-tooling-docs] | Reference | What It Covers | Use It For | | ------------------------------------------------------------------------ | ------------------------------------------------------------------------------------------------- | ------------------------------------------------------------ | | [Mastra Docs](https://mastra.ai/docs) | TypeScript agents, tools, workflows, memory, RAG, MCP, evals, and voice | Default local TypeScript agent scaffolding. | | [Vercel AI SDK](https://ai-sdk.dev/docs) | Model calls, tools, structured output, streaming, and UI | TypeScript AI app foundations. | | [OpenAI Agents SDK](https://platform.openai.com/docs/guides/agents-sdk) | Agents, tools, handoffs, guardrails, tracing, and evals | OpenAI-native agent systems. | | [OpenRouter Docs](https://openrouter.ai/docs) | Model routing, structured outputs, plugins, and hosted tools | Provider optionality and first-party model-call tools. | | [Cloudflare Agents](https://developers.cloudflare.com/agents/) | Durable Object-backed agents, state, scheduling, tools, MCP, and browser access | Workers-native agent systems. | | [Cloudflare AI Gateway](https://developers.cloudflare.com/ai-gateway/) | Gateway observability, caching, rate limits, guardrails, retries, and routing | Governed model access across providers. | | [Cloudflare Browser Run](https://developers.cloudflare.com/browser-run/) | Managed browser automation, screenshots, PDFs, Markdown extraction, crawling, Playwright, and MCP | Browser access for agents. | | [Cloudflare Sandbox](https://developers.cloudflare.com/sandbox/) | Isolated code execution for agents and applications | Running untrusted or generated code outside the app process. | ## Commerce And Payments [#commerce-and-payments] | Reference | What It Covers | Use It For | | ------------------------------------------------------------- | ------------------------------------------- | -------------------------------------------------------- | | [x402](https://www.x402.org/) | HTTP-native machine payments | Pay-to-access resources and APIs. | | [Universal Commerce Protocol](https://ucp.dev/) | Commerce discovery and transaction protocol | Agent-driven ecommerce surfaces. | | [Agentic Commerce Protocol](https://www.agenticcommerce.dev/) | Agent-mediated purchasing flows | Agent-initiated buying with explicit payment boundaries. | ## How This Page Relates To The Catalog [#how-this-page-relates-to-the-catalog] Use this page for canonical sources and specs. Use the [Tooling Catalog](/docs/tooling-catalog) for curated recommendations, tiers, and judgement calls.