Recipes

Named cross-category compositions: real systems compose many patterns at once. Each recipe lists the patterns that together produce a recognisable kind of system. Members carry a role: 'core' patterns define the recipe and removing one breaks the shape; 'hardening' patterns add safety, observability, or correctness; 'optional' patterns are common variations or upgrades.

Modern Coding Agent

An agent that reads, writes, and runs code in a sandbox, calling tools and (optionally) sub-agents while a human approves the destructive parts. The shape that powers Cursor, Claude Code, OpenHands, Aider, Codex CLI.

Production RAG

Retrieval-grounded generation built to be defensible: hybrid retrieval, reranking, contextualised chunks, citations rendered to the user, and verification before the answer ships.

Voice Agent Stack

A low-latency conversational agent over a phone or microphone, with handoff to humans, mid-utterance cancellation, and per-call session boundaries. The shape behind LiveKit, Pipecat, Vapi, Retell.

Sovereign / Regulated Deployment

An agent stack that satisfies data-residency and audit requirements: weights, inference, tools, and logs all sit inside an operator-controlled boundary, with provenance and incident response wired in.

Long-Running Autonomous Agent

An agent that operates over hours to weeks, surviving restarts and accumulating memory while remaining safe. The shape behind Devin, Manus, Sparrot, durable LangGraph runs.

Multi-Agent Debate

Two or more agents argue toward a better answer than any single agent would produce, with a frozen rubric to score the result. The shape behind debate-style alignment work and 'committee of critics' setups.

Browser & Computer-Use Stack

An agent that drives a real GUI: planning a task, grounding actions in pixels or DOM, and asking permission before destructive clicks. The shape behind OpenAI Operator, Anthropic Computer Use, Browser Use, Stagehand, MultiOn.

Memory Architecture

How long-running agents structure what they remember: tiered short-to-long-term cascade, compaction across the window, paging, and reasoning carry-forward across tool calls.

Multi-Agent Coordination

Several agents collaborate under a coordinator, with explicit hand-offs and a shared protocol. The shape behind LangGraph supervisor, OpenAI Swarm, AutoGen group chat, Bedrock multi-agent orchestrators.

Safety Hardening

The minimum set of constraints to put around any production agent before it touches the world: budgets, gates, charters, kill-switches, approvals.

Eval & Observability

How you keep an agent honest in production: harness, judge, decision log, provenance, shadow rollouts.

Structured Output Stack

Get typed, schema-conformant data out of the model and verify it. The shape behind Outlines, Instructor, Pydantic AI, DSPy.

Streaming UX Stack

User-perceivable real-time output: tokens streamed as they arrive, citations attached as they resolve, the user can stop at any time and the agent can interrupt the user when something matters.

Planning Loops

Different ways to structure 'think then act': linear ReAct, plan-then-execute, parallel DAG planning, tree search with backtracking, and the outer/inner planner+executor split.

Routing & Fallback

How requests get to the right model or specialist and how the system stays up when one upstream breaks. The shape behind LangChain fallbacks, model routers, provider cascades.

Reflection & Self-Correction

Patterns where the model reviews its own work before shipping it: scoped rubric reflection, self-refine, deterministic post-checks, process rewards.