ChangeGamer

Guides

Long-form guides in 12 topic clusters. Each cluster is a pillar guide plus in-depth sub-articles, and every guide is free.

153 articles · 260,821 words · JSON index · Atom feed · JSON feed

Guides vs resources. These guides are written for the person building or running a site: decisions, trade-offs and implementation. The reference corpus is written for machines consuming a contract — formats, protocols, token tables, payment flows. Guide writers must trace each factual claim to a reference page or to a source named in the text, and every guide has a .md variant for agents.

Recently updated: Agentic Browsing in Production: Building, Securing and Preparing for Browser Agents () · OSWorld vs WebArena: Which Benchmark Tells You Anything About Your Browser Agent? () · How to Add WebMCP Tools to Your Website () · Headless Browser for AI Agents: Local, Remote or Managed? () · How to Sign Browser Agent Requests With Web Bot Auth ()

The agent-ready web

How to make a website readable, citable, controllable and payable for AI agents and AI crawlers — the operator side of the machine-first web.

13 of 13 sub-articles · complete · started

Selling to AI agents

How to sell content, APIs and tools to buyers that are software — discovery, machine-readable offers, spend ceilings, payment rails, and what breaks.

12 of 12 sub-articles · complete · started

MCP in practice

How to build, ship and run an MCP server in production — transport, auth, tool design, versioning, testing, distribution, observability, cost and failure modes.

12 of 12 sub-articles · complete · started

RAG in production

How to run retrieval-augmented generation as a real system — ingestion and chunking, embedding choice and reindexing, hybrid search, reranking, evaluation, freshness, access control, cost, latency, and when RAG is the wrong answer.

13 of 13 sub-articles · complete · started

Agent reliability in production

How to make an AI agent reliable — tool-calling contracts, structured outputs, retries and idempotency, timeouts, durable execution, guardrails, evaluation in CI, observability, incident response, and rollout.

12 of 12 sub-articles · complete · started

Agent security operations

How to defend an AI agent deployment from the operator side — secrets and credential hygiene, prompt-injection defense in depth, sandboxing, supply-chain provenance, least privilege, audit trails, incident response, and rate/abuse controls — not buyer-side fraud and not reliability-framed guardrails.

12 of 12 sub-articles · complete · started

Agent observability and evaluation

How to observe and evaluate an AI agent already live in production — tracing spans for tool calls and retrieval steps, judge-based screening versus human acceptance of live output, online metrics, cost telemetry, and incident postmortems that feed evaluation fixtures — not the pre-ship, CI-gating evaluation methodology already owned by evaluating-ai-agents-in-ci, and not the trace/span field mechanics already owned by agent-observability-for-reliability, both in the completed agent-reliability cluster.

11 of 11 sub-articles · complete · started

Voice agents in production

How to build and run a live voice AI agent — architecture choice, turn detection and barge-in, latency budget, tool calls in a voice loop, transport, cost, human handoff, guardrails, monitoring and launch — covering live caller conversations only, not offline dubbing/localization (ai-dubbing-localization-and-subtitling) and not the WER/MOS/VoiceBench scoring methodology owned by the evaluating-voice-agents resource.

12 of 12 sub-articles · complete · started

Support agents in production

How to run an AI customer support agent in production — escalation triggers and warm handoff, helpdesk integration, refund-policy enforcement, tone and brand voice, customer memory, CSAT and quality measurement, human review queues, multichannel deployment, wrong-answer incidents and replay testing — covering the support vertical only, not the horizontal guardrail and security mechanics owned by agent-guardrails-for-reliability and not the voice mechanics owned by voice-agents-in-production.

12 of 12 sub-articles · complete · started

Document extraction pipelines in production

How to run AI document extraction as a production pipeline — OCR vs vision-model vs hybrid parsing, schema-constrained field extraction, table and form validation, human-review fallback, schema versioning, cost per page, open-source vs managed tooling and feeding a system of record — not chunking and retrieval (chunking-documents-for-rag, rag-ingestion-pipeline), not generic structured-output and tool-call mechanics (structured-outputs-vs-tool-calling, tool-calling-contracts-for-ai-agents) and not injection defense in depth (prompt-injection-defense-in-depth-for-agents).

11 of 11 sub-articles · complete · started

A2A in practice: agent-to-agent interoperability

How to build, secure and run agent-to-agent delegation with the A2A protocol — when A2A beats in-process subagents, Agent Cards, task lifecycle, bindings, authentication, credential propagation across hops, discovery and trust, v0.3 to v1.0 migration, framework support and failure handling — not MCP tool serving (mcp-in-practice) and not single-agent identity (agent-identity-and-authentication-for-ai-agents).

12 of 12 sub-articles · complete · started

Browser agents and agentic browsing in production

How to build, run and prepare for browser-driving agents — the computer-use loop, accessibility-tree vs screenshot grounding, cost and latency, page-borne prompt injection, logged-in-session risk, remote browsers and anti-bot, Web Bot Auth request signing, WebMCP for site owners, evaluation and the product landscape — not crawler and retrieval readiness (agent-ready-web) and not generic prompt-injection defense (agent-security-operations).

9 of 11 sub-articles · in progress · started