Methodology & Bibliography
How this site is sourced, which canonical primary sources back its claims, and how the data is refreshed.
Data last refreshed: 2026-05-08. Next scheduled refresh: Q4 2026, or sooner on regulator/vendor trigger (A2A spec revision; IETF AAuth RFC progression; NIST CAISI publication; Microsoft Entra Agent ID major update).
Sourcing rule
Every claim has at least one direct quote or paraphrase from a canonical primary source — vendor docs, standards bodies, regulator publications, peer-reviewed papers, or an analyst publication where the analyst is the primary source for the framing. Wikipedia and secondary blogs are never the primary citation. When a topic is contested, both positions are carried with attributed flags.
Freshness flags
- Foundational — original peer-reviewed papers (e.g., Mitchell et al. 2019 Model Cards). Refreshed only if substantively revised.
- In force — current binding regulator text or in-effect specifications, with the in-force date.
- Emerging 2024–2026 — terms or specs that entered mainstream discourse recently and are still evolving.
- Contested — entries with named, meaningful disagreement in the field; we carry both positions.
Research sprint
This site was first promoted to tier-1 enrichment on 2026-05-08. The full audit trail — sources surveyed, quotable lines captured, stale claims retired, open questions for the next sprint — lives in the repository at raw/research-notes.md. The sprint covered the eight research dimensions (R1 vendor-canonical updates, R2 regulator activity, R3 analyst output, R4 academic frontier, R5 competitor moves, R6 search-intent signals, R7 fresh quotables, R8 counter-narratives) per the build-authority-property workflow.
Refresh cadence
- Default: bi-annual full review (per the SKILL Part G mechanic table for pillar mini-sites).
- Trigger refresh: A2A spec revision (v1.1+); IETF AAuth RFC progression; NIST CAISI publication; Microsoft Entra Agent ID major update; new Linux Foundation founding-member announcement; major marketplace launch (Salesforce / Google / AWS / Microsoft).
- Citation rot check: quarterly URL-liveness scan over every cited source.
Bibliography
- A2A Protocol — Specification (v1.0). a2a-protocol.org/latest/specification/
- Google Developers — Announcing the Agent2Agent Protocol. developers.googleblog.com
- Linux Foundation — Launches the Agent2Agent Protocol Project. linuxfoundation.org press release
- github.com/a2aproject/A2A — reference implementation, JSON schemas, TypeScript types. github.com/a2aproject/A2A
- A2A Project — Partners. a2a-protocol.org/latest/community/partners/
- Mitchell et al., Model Cards for Model Reporting (arXiv:1810.03993). arxiv.org/abs/1810.03993
- RFC 9309 — Robots Exclusion Protocol. rfc-editor.org/rfc/rfc9309.html
- llms.txt — proposal (Howard, Answer.AI, Sept 2024). llmstxt.org
- OpenAPI Specification 3.1.0. spec.openapis.org/oas/v3.1.0.html
- Model Context Protocol — Specification (2025-11-25). modelcontextprotocol.io/specification/2025-11-25
- Anthropic — Introducing the Model Context Protocol. anthropic.com/news/model-context-protocol
- MCP Blog — The 2026 MCP Roadmap. blog.modelcontextprotocol.io/posts/2026-mcp-roadmap/
- Microsoft — Announcing Microsoft Entra Agent ID. techcommunity.microsoft.com
- Microsoft — Agent Framework v1.0. devblogs.microsoft.com/agent-framework
- Microsoft Learn — Agent Framework Overview. learn.microsoft.com/en-us/agent-framework/overview/
- Salesforce — AgentExchange launch (March 2025). salesforce.com press release
- Salesforce — Agentforce A2A integration with Google Cloud. salesforce.com press release
- Google Cloud — Vertex AI Agent Builder / Agentspace. cloud.google.com/products/agent-builder
- Google — Agent Development Kit (ADK). google.github.io/adk-docs/
- AWS — Bedrock AgentCore Runtime. docs.aws.amazon.com/bedrock-agentcore
- NIST CAISI — AI Agent Standards Initiative. nist.gov/caisi/ai-agent-standards-initiative
- NIST — Announcing the AI Agent Standards Initiative. nist.gov/news-events/news/2026/02
- IETF — AAuth, Agentic Authorization OAuth 2.1 Extension (draft-rosenberg-oauth-aauth-00). ietf.org/archive/id/draft-rosenberg-oauth-aauth-00.html
- OWASP — Top 10 for Large Language Model Applications. owasp.org/www-project-top-10-for-large-language-model-applications/
- Gartner — Hype Cycle for Agentic AI 2026. gartner.com/en/articles/hype-cycle-for-agentic-ai
- McKinsey QuantumBlack — The state of AI in 2025 (Nov 2025). mckinsey.com
- arXiv:2601.13671 — Orchestration of Multi-Agent Systems. arxiv.org/abs/2601.13671
- Anthropic — Building Effective Agents. anthropic.com/research/building-effective-agents
- Marcus, G. — Six (or seven) predictions for AI 2026. garymarcus.substack.com
Audit trail: raw/research-notes.md in this satellite's source repository.