References

Every source cited across the topic catalog, consolidated and alphabetized. Sources are primary documentation, standards bodies, peer-reviewed or preprint research, and named engineering organizations' own published data: not marketing content, SEO content, or generic consultant thought leadership. This list is a lookup appendix; the in-context citation for each source appears under the relevant topic's "Recommended Learning" entry in the catalog.

• A Careful Examination of Large Behavior Models for Multitask Dexterous Manipulation (arXiv)
• A Gentle Introduction to Conformal Prediction and Distribution-Free Uncertainty Quantification: Angelopoulos and Bates (arXiv)
• A Timeline of Model Context Protocol (MCP) Security Breaches
• A2A and MCP: A2A Protocol
• A2A Protocol Surpasses 150 Organizations...: Linux Foundation
• Access Foundry Models and Other Language Models Through a Gateway: Azure Architecture Center
• Activating your internal AI champions (GitHub)
• Against Predictive Optimization: Wang, Kapoor, Barocas, Narayanan (SSRN)
• Agent Authority Least Privilege Framework: FINOS
• Agentic AI: Threats and Mitigations
• AgentOps: Enabling Observability of LLM Agents (arXiv)
• AgentPoison: Red-teaming LLM Agents via Poisoning Memory or Knowledge Bases
• Agents, robots, and us: Skill partnerships in the age of AI: McKinsey Global Institute
• agents.md: the open standard
• Agility Robotics reports $1.8M revenue ahead of humanoid SPAC: The Robot Report
• AI Act: Shaping Europe's digital future (European Commission)
• AI Agents as Users: Nielsen Norman Group
• AI Agents Have an Authorization Problem, Not Just an Identity Problem
• AI Chatbots Discourage Error Checking: Nielsen Norman Group
• AI gateway capabilities in Azure API Management
• AI Red Teaming Initiative: OWASP Gen AI Security Project
• AI Risk Management Framework (NIST)
• AI RMF: Generative Artificial Intelligence Profile (NIST)
• Amazon deploys its 1 millionth robot, releases generative AI model: TechCrunch
• Announcing Microsoft Entra Agent ID
• Anthropic: Claude Code best practices
• Anthropic: Effective harnesses for long-running agents
• Anthropic: How we built our multi-agent research system
• appleinsider.com
• Article 4: AI literacy: EU Artificial Intelligence Act
• Artificial Intelligence Risk Management Framework (AI RMF 1.0), NIST AI 100-1
• Auditing and Logging AI Agent Activity: A Guide for Engineers: LoginRadius Engineering Blog
• Authorization and Governance for AI Agents: Runtime Authorization Beyond Identity at Scale
• Bersin
• BIRD-SQL benchmark
• Block denied topics to help remove harmful content: Amazon Bedrock
• Building Effective AI Agents: Anthropic
• Canaries Dashboard (Stanford Digital Economy Lab)
• Canaries in the Coal Mine? Six Facts about the Recent Employment Effects of AI (Stanford Digital Economy Lab)
• Canaries in the Coal Mine? Six Facts about the Recent Employment Effects of Artificial Intelligence: Stanford SIEPR
• Chargeback vs. Showback: Cloud Cost Allocation Models Explained: CloudZero
• ChatGPT Outperforms Crowd Workers for Text-Annotation Tasks: Gilardi, Alizadeh, Kubli, PNAS
• Cisco Survey Reveals Close to Three-Fourths of IoT Projects Are Failing: Cisco
• Claude Enterprise
• Claude prompting best practices (Claude Platform Docs)
• Cloud Security Alliance: The Vibe Coding Governance Gap
• cloud.google.com
• Community Control Over Police Surveillance: ACLU
• Computer use tool: Claude Platform Docs
• Computer-Using Agent: OpenAI
• Contextual Retrieval (Anthropic)
• Decision Models vs. LLMs as Rubric Judges: Cheaper, Faster, and Wrong in the Same Places: Rao and Callison-Burch (arXiv)
• Decisions: OpenAI API Documentation
• deloitte.com
• Deloitte: The State of AI in the Enterprise
• Deploy and Use Microsoft-Decision-1 in Microsoft Foundry: Microsoft Learn
• developers.googleblog.com
• Directive on Automated Decision-Making: Treasury Board of Canada Secretariat
• Dissecting Racial Bias in an Algorithm Used to Manage the Health of Populations: Obermeyer and coauthors, Science
• Diving Into the MCP Authorization Specification: Descope
• DoD Directive 3000.09, Autonomy in Weapon Systems: US Department of Defense
• Donating the Model Context Protocol and establishing the Agentic AI Foundation: Anthropic
• DORA 2025: State of AI-assisted Software Development
• DORA AI Capabilities Model report
• DORA Insights
• DORA research on AI
• DORA: Balancing AI tensions: moving from adoption to effective SDLC use
• Effective context engineering for AI agents: Anthropic
• Embodied AI: China's ambitious path to transform its robotics industry: MERICS
• en.wikipedia.org
• Enterprise AI Operating Model: Hub-and-Spoke, Federated, or Centralized?
• EU agrees Digital Omnibus deal to simplify AI rules: White & Case
• EU AI Act implementation timeline: AI Act Service Desk
• Explainable AI in Chat Interfaces: Nielsen Norman Group
• Factory.ai: From coding agents to software factories
• FCC Adds All Foreign-Produced Advanced Robotic Devices to the Covered List: Sidley Austin
• Fine-tuning (OpenAI Developers guide)
• FinOps for AI Overview: FinOps Foundation
• Five million robots now operate in factories globally: International Federation of Robotics
• Forrester
• FrugalGPT: How to Use Large Language Models While Reducing Cost and Improving Performance: Chen, Zaharia, Zou (arXiv)
• Gartner
• Gartner Predicts 40% of Enterprise Apps Will Feature Task-Specific AI Agents by 2026
• gartner.com
• Gemini 3: Introducing the latest Gemini AI model (Google)
• Generative AI at Work: Brynjolfsson, Li, Raymond (NBER Working Paper 31161)
• Generative Artificial Intelligence and Copyright Law: Congressional Research Service
• GitClear: AI Copilot Code Quality 2025 research
• GitHub MCP Exploited: Accessing private repositories via MCP
• GitHub Octoverse 2025
• GitHub Octoverse 2025
• Glean
• Glean
• Global sales of professional service robots surge 24 percent: International Federation of Robotics
• Google Cloud
• Google: NotebookLM Adds Deep Research
• GPT-5.1 System Card Addendum (OpenAI)
• GraphRAG: Unlocking LLM discovery on narrative private data (Microsoft Research)
• Healthcare.gov: Ineffective Planning and Oversight Practices Underscore the Need for Improved Contract Management (GAO-14-694): US Government Accountability Office
• HHEM v2: A New and Improved Factual Consistency Scoring Model: Vectara
• Hidden Technical Debt in Machine Learning Systems: Sculley and coauthors, NeurIPS
• How we built our multi-agent research system: Anthropic
• Hybrid search (Weaviate Documentation)
• Identifying Token Costs Hiding in Your Agentic Loop: MachineLearningMastery
• InfoQ: Agentic fitness functions: extending evolutionary architecture beyond deterministic rules
• InfoWorld: How Google is using LLMs for complex internal code migrations
• Inside the LLM Call: GenAI Observability with OpenTelemetry: OpenTelemetry Blog
• Introducing Claude Opus 4.5 (Anthropic)
• Introducing Clef: Cloudflare's First Open-Source Decision Models, Now on Workers AI: Cloudflare
• Introducing System One Models & Jev: TypeSafe AI
• Introducing the Model Context Protocol (Anthropic)
• Introduction: Model Context Protocol
• ISO/CD 25785-1 (ISO)
• ISO/IEC 42001 explained
• ISO/IEC 42001:2023: AI management systems
• ISO/IEC 42001:2023: Microsoft compliance overview
• Jev 1.13 Jaggedness: TypeSafe AI documentation
• JEV versus LLMs: Accuracy, Cost and Calibration on Seven Political Science Replications: Denney and DiGiuseppe (arXiv)
• Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena (arXiv)
• Komatsu becomes first OEM to commission 1,000 ultra-class autonomous haul trucks: Komatsu
• Kroger to close three automated delivery fulfillment centers: The Spokesman-Review
• Language Models (Mostly) Know What They Know: Kadavath and coauthors (arXiv)
• Large Language Models Sensitivity to the Order of Options in Multiple-Choice Questions: Pezeshkpour and Hruschka (arXiv)
• learn.microsoft.com
• Least privilege for AI agents with Microsoft Entra Agent ID
• Least privilege for AI agents: Identity, access, and tool binding
• Levels of Autonomy for AI Agents: Knight First Amendment Institute
• Leveraging model distillation to fine-tune a model (OpenAI Cookbook)
• Liability for defective products: European Commission
• Llama 4 Community License Agreement
• LLM Agent Evaluation Metrics: Tool Calling, Task Completion, Reasoning, and Trace-Based Evals: Confident AI
• LLM Evaluation: Best Practices and Methods: Databricks Engineering Blog
• LLM Routing and Model Cascades: How to Cut AI Costs Without Sacrificing Quality
• LLMOps Guide: Build Fast, Cost-Effective LLM Apps: Redis Engineering Blog
• MCP Authorization specification
• MCP Security Notification: Tool Poisoning Attacks: Invariant Labs
• MCP Tool Poisoning: OWASP Foundation
• Measuring AI agent autonomy in practice: Anthropic
• MemGPT: Towards LLMs as Operating Systems (arXiv)
• Memory for Claude Managed Agents (Anthropic)
• merics.org
• Meta's Llama license is still not Open Source (Open Source Initiative)
• METR: Measuring the impact of AI on experienced open-source developer productivity
• Microsoft Agent 365 Overview
• Microsoft GraphRAG project page
• Microsoft Learn: Overview of Process Mining in Power Automate
• Microsoft Support: Use Researcher in Copilot Notebooks
• Microsoft: Catch Up on Meetings with Copilot in Teams
• Migrating Code At Scale With LLMs At Google (arXiv 2504.09691, FSE 2025)
• MIT NANDA
• MIT report: 95% of generative AI pilots at companies are failing
• Model router for Microsoft Foundry: concepts
• n8n Docs
• NIST AI 600-1: Artificial Intelligence Risk Management Framework: Generative AI Profile
• NIST AI RMF Playbook
• Not What You've Signed Up For: Compromising Real-World LLM-Integrated Applications with Indirect Prompt Injection: Greshake and coauthors (arXiv)
• NSA/CISA: Model Context Protocol (MCP) Security
• On Calibration of Modern Neural Networks: Guo, Pleiss, Sun, Weinberger (arXiv)
• One Year of MCP (Model Context Protocol blog)
• Open X-Embodiment: Robotic Learning Datasets and RT-X Models (arXiv)
• OpenAI
• OpenTelemetry GenAI Semantic Conventions: MLflow documentation
• OSWorld 2.0 (arXiv 2606.29537)
• Overview of model routing: Google Cloud API Gateway
• OWASP Gen AI Security Project: Resources
• OWASP LLM01:2025 Prompt Injection
• OWASP LLM04:2025 Data and Model Poisoning
• OWASP LLM05:2025 Improper Output Handling
• OWASP LLM06:2025 Excessive Agency
• OWASP LLM08:2025 Vector and Embedding Weaknesses
• OWASP Top 10 for LLM Applications 2025
• OWASP Top 10 for LLM Applications 2025
• pgvector (GitHub)
• Prediction Policy Problems: Kleinberg, Ludwig, Mullainathan, Obermeyer, American Economic Review
• Prediction-Powered Inference: Angelopoulos, Bates, Fannjiang, Jordan, Zrnic (arXiv)
• Predictions Scorecard, 2026 January 01: Rodney Brooks
• Prompt engineering best practices for 2026 (Anthropic)
• Prompt injection: Wikipedia
• RAG Security Cheat Sheet (OWASP)
• RAG with Permissions (Supabase Docs)
• RAGAS: Automated Evaluation of Retrieval Augmented Generation (arXiv)
• ReAct: Synergizing Reasoning and Acting in Language Models
• Regulation (EU) 2024/1689 (EUR-Lex, official text)
• Releasing the Inner Jev: A Simple Recipe for Open Decision Models (arXiv)
• Remove PII from conversations by using sensitive information filters
• Replay: reconstructing an agent action from the audit log: Deixic
• research.google
• Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks (Lewis et al., 2020)
• Rising AI Adoption Spurs Workforce Changes (Gallup)
• Robots and Jobs: Evidence from US Labor Markets: Acemoglu and Restrepo, MIT DSpace
• RouteLLM: Learning to Route LLMs with Preference Data: Ong and coauthors (arXiv)
• RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Control (arXiv)
• Safety and content filters: Vertex AI
• salesforce.com
• salesforce.com
• Salesforce: Agentforce Developer Guide: Get Started
• Securing Agentic Applications Guide 1.0
• Security Degradation in Iterative AI Code Generation (arXiv 2506.11022, IEEE ISTAS 2025)
• Security, Guardrails, and Observability in Amazon Bedrock
• Shadow deployment vs. canary release of machine learning models
• Simon Willison's prompt-injection archive
• Smart Streetlights: Voice of San Diego
• SPIFFE: Securing the identity of agentic AI and non-human actors
• State of UX 2026: Nielsen Norman Group
• Stop Explaining Black Box Machine Learning Models for High Stakes Decisions and Use Interpretable Models Instead: Rudin, Nature Machine Intelligence
• stripe.com
• Structured Outputs (OpenAI API)
• Team Topologies as the "infrastructure for agency" with AI
• Team Topologies: the book's site
• Techzine
• The 2026 MCP Roadmap: Model Context Protocol Blog
• The Adjustment of Labor Markets to Robots: Dauth and coauthors, University of Konstanz
• The AI Act Implementation Timeline: What Changes Under the AI Omnibus: Future of Privacy Forum
• The AI Champion role (OpenAI Academy)
• The lethal trifecta for AI agents
• The ML Test Score: A Rubric for ML Production Readiness and Technical Debt Reduction: Breck and coauthors, Google Research
• The New York Times v. Microsoft and OpenAI: case background and status
• The Non-Human Identity Governance Vacuum
• The Register, "AI coding tools make developers slower, study finds"
• The Simple Economics of Machine Intelligence: Harvard Business Review
• The State of AI: Global Survey 2025 (McKinsey)
• The Vulnerable MCP Project
• Token Economics: The Atomic Unit of AI Value: FinOps Foundation
• Tool use with Claude: Claude Platform Docs
• TRAJECT-Bench: A Trajectory-Aware Benchmark for Evaluating Agentic Tool Use (arXiv)
• UAS Drone Program: City of Chula Vista Police Department
• UiPath
• UiPath: What is Agentic Orchestration?
• Understanding intelligent prompt routing in Amazon Bedrock
• Understanding the context window (Anthropic docs)
• US Supreme Court declines to consider whether AI alone can create copyrighted works: Morgan Lewis
• vectara/hallucination-leaderboard (GitHub)
• Waymo is scaling fast: Here's what the fleet data shows: TechCrunch
• Waymo skyrocketing ridership in one chart: TechCrunch
• When Should We Trust AI? Magic-8-Ball Thinking and AI Hallucinations: Nielsen Norman Group
• Why Today's Humanoids Won't Learn Dexterity: Rodney Brooks
• Workato: Agentic (docs)
• Zipline Surpasses 2 Million Deliveries, Raises More than $600M: Zipline