# Agents Field Guide > Prompts, methods, and worked examples for building with agents. Full documentation (discoverable current pages, one document): https://afg.callum.website/llms-full.txt ## Pages - [Agents Field Guide](https://afg.callum.website/index.md) — Prompts, methods, and worked examples for building with agents. - [AgentsView learning](https://afg.callum.website/agentsview-learning/index.md) - [Ask for proof of work instead of a summary](https://afg.callum.website/ask-for-proof-of-work/index.md) - [Automate agreed rules and surface open decisions](https://afg.callum.website/automate-agreed-rules/index.md) — Give the agent checks it can act on and reviewers the decisions that change what callers can rely on. - [Take a breadboard from sketch to proof](https://afg.callum.website/breadboard-to-proof/index.md) — Map an upload, ask an agent to build it, and prove the result survives reopening. - [The Case Against Formal Verification, 50 Years Later](https://afg.callum.website/case-against-formal-verification/index.md) - [Coordinate work in Codex](https://afg.callum.website/coordinate-work-in-codex/index.md) — Split work across agents and chats, and keep a shared worklog. - [An eval with held-out support tickets](https://afg.callum.website/eval-with-held-out-support-tickets/index.md) - [External skills](https://afg.callum.website/external-skills/llms.txt) - [Field reports](https://afg.callum.website/field-reports/index.md) — What I’m trying, what held up in practice, and what changed. - [Find implicit state machines with Pstack](https://afg.callum.website/find-implicit-state-machines/index.md) - [Four Dark Corners of Requirements Engineering](https://afg.callum.website/four-dark-corners-of-requirements-engineering/index.md) — How a software specification, together with facts about the world, delivers a requirement. - [Four design options side by side](https://afg.callum.website/four-design-options-side-by-side/index.md) - [Intent specialists](https://afg.callum.website/give-a-job-to-a-specialist/index.md) — Six reusable agent roles and when to use them. - [Give a long task a worklog](https://afg.callum.website/give-a-long-task-a-worklog/index.md) — Keep decisions and open questions that would otherwise disappear with the chat. - [Give the agent a way to verify its work](https://afg.callum.website/give-the-agent-a-way-to-verify/index.md) - [Work in independently valuable steps](https://afg.callum.website/independently-valuable-steps/index.md) - [Playbooks](https://afg.callum.website/methods/index.md) — Ways to shape work and give agents a way to check it. - [Prompts](https://afg.callum.website/prompts/llms.txt) - [Prove the thing the user can actually do](https://afg.callum.website/prove-the-user-outcome/index.md) — Turn the user’s promise into an observable check that can catch a plausible failure. - [Get an agent unstuck when it keeps missing the point](https://afg.callum.website/recover-drifting-work/index.md) - [References](https://afg.callum.website/references/index.md) — Articles and papers I return to. - [Skills I publish](https://afg.callum.website/skills-i-publish/index.md) - [Skills](https://afg.callum.website/skills/index.md) - [T3 Code with Opus and DeepSeek](https://afg.callum.website/t3-code-experiment/index.md) - [ToDesktop’s visual system docs](https://afg.callum.website/todesktop-visual-system-docs/index.md) - [Tools I use](https://afg.callum.website/tools/index.md) — Tools I use for screen feedback and dictation. - [The Topology of an LLM Agent Loop With Tool Use](https://afg.callum.website/topology-of-an-llm-agent-loop/index.md) - [Towards Self-Driving Codebases](https://afg.callum.website/towards-self-driving-codebases/index.md) - [Turn repeatable agent work into a script](https://afg.callum.website/turn-repeatable-agent-work-into-a-script/index.md) — Use the agent to work out the rules, then script the steps you can check. - [Describe how an existing product behaves](https://afg.callum.website/understand-an-existing-product/index.md) — Trace what people experience when a flow is interrupted. - [Why I turned off automatic agent memory](https://afg.callum.website/why-i-turned-off-automatic-agent-memory/index.md) — I tested whether lessons saved from old chats helped agents on four new tasks.