From a rough idea to a production agent system.
Explore the full lifecycle: agents, Skills, architecture, design, specifications, plans, tasks, memory, context, tools, multi-agent work, evals, safety, operations, governance, commerce, and more. Every topic exposes its decision rule, artifacts, questions, and primary evidence.
21
maintained topics
27
unique primary sources
7 days
new brief cadence
Content engine
New regularly. Updated carefully.
Automation checks official sources, validates every topic, and rotates six topics into a new source-linked brief each week. If a source breaks or the review window expires, publishing stops and a review issue opens instead of silently generating unsupported claims.
- 01
Check sources
Verify review age, structure, internal routes, and live primary documentation.
- 02
Publish a rotation
Create a deterministic weekly brief from six reviewed topics and their evidence.
- 03
Block stale content
Open a review issue and fail the workflow when evidence or freshness needs attention.
Topic library
Follow the lifecycle or jump to a decision.
Foundations
What the system is and how its responsibilities are divided.
Agents
Agents combine models, instructions, tools, state, and an execution loop so the system can decide how to pursue an outcome. The important design choice is not whether to use the label agent, but how much control the model receives and where deterministic boundaries remain.
Read the field guideSkills
Skills package reusable instructions, examples, scripts, and resources for a repeatable workflow. They should stay focused enough to load only when relevant and concrete enough that another agent can follow the procedure without rediscovering it.
Read the field guideAgent architecture
Agent architecture assigns responsibility across models, deterministic code, tools, state stores, queues, human reviewers, and independently operated agents. A useful architecture makes control flow, trust boundaries, and recovery behavior visible before framework details.
Read the field guideDelivery
How intent becomes designs, specifications, plans, and tasks.
Agent product design
Agent product design turns uncertain model behavior into an understandable user experience. It defines what the agent can do, how it communicates progress and uncertainty, when users can inspect or interrupt work, and what recovery looks like after failure.
Read the field guideSpecifications
Specifications convert intent into reviewable requirements, acceptance scenarios, constraints, and non-goals. For agent-assisted delivery, a specification should be precise enough to guide planning and validation without pretending that uncertain decisions are already settled.
Read the field guidePlans
A plan translates approved intent into technical decisions, dependencies, sequencing, migration strategy, and verification. Strong plans preserve the why behind choices and identify uncertainty before work is fragmented into tasks.
Read the field guideTasks
Tasks turn a plan into bounded units with inputs, outputs, dependencies, acceptance checks, and ownership. Agent-ready tasks minimize hidden context and produce results that can be verified and integrated without replaying the entire reasoning process.
Read the field guideRuntime
How context, tools, agents, protocols, and workflows execute.
Memory and context
Context is the information available for the current inference; memory is persisted state that may shape later runs. Reliable systems separate conversation history, durable facts, user preferences, project policy, and derived summaries so each can be governed and refreshed appropriately.
Read the field guideContext engineering
Context engineering controls which instructions, examples, tools, retrieved facts, messages, and summaries reach the model at each step. The objective is not maximum context; it is the smallest current set that makes the desired behavior likely and reviewable.
Read the field guideTools and MCP
Tools let models request deterministic functions; MCP standardizes how compatible hosts discover and call external tools, resources, and prompts. Good tool design minimizes ambiguity, constrains inputs, makes side effects visible, and returns results the model can interpret reliably.
Read the field guideMulti-agent systems
Multi-agent systems divide work among specialized or isolated agents and synthesize their outputs. They are valuable when work partitions cleanly or trust boundaries differ, but they add coordination, context transfer, conflict resolution, and observability costs.
Read the field guideAgent interoperability
Interoperability allows hosts, tools, and independently operated agents to discover capabilities and exchange work through explicit contracts. MCP primarily connects models to tools and context; A2A primarily coordinates agents and task artifacts across service boundaries.
Read the field guideWorkflows and automation
Workflows encode known control paths; automations trigger those workflows on a schedule or event without continuous human initiation. Reliable automation requires idempotency, retry policy, durable state, ownership, observability, and a clear boundary for escalation.
Read the field guideOperations
How quality, telemetry, releases, and recovery stay accountable.
Evals and verification
Evals measure behavior across representative tasks; verification checks whether a specific run produced an acceptable result. Production systems need both: aggregate evidence for change decisions and concrete completion checks before outputs or side effects are trusted.
Read the field guideObservability
Agent observability connects user outcomes to model turns, tool calls, handoffs, guardrails, latency, token use, and errors. Useful telemetry makes an end-to-end workflow explainable while controlling sensitive prompt, output, argument, and result data.
Read the field guideDeployment and operations
Operating agents in production requires durable run identities, bounded time and cost, resumable state, safe retries, versioned dependencies, rollback, observability, and ownership. Deployment is a change to a live decision system, not merely a code upload.
Read the field guideGovernance
How authority, safety, review, and organizational policy are enforced.
Safety and permissions
Safety combines behavioral guidance with enforceable controls around data, tools, networks, execution, and human approval. Prompts can shape intent, but least-privilege permissions and sandbox boundaries limit what mistakes or attacks can actually affect.
Read the field guideHuman approval
Human approval is a control boundary for decisions that require accountable judgment. A useful approval presents the proposed action, arguments, evidence, consequences, and alternatives, then persists the decision so a paused run can resume safely.
Read the field guideGovernance
Agent governance assigns ownership and policy across instructions, models, tools, data, evaluation, permissions, releases, incidents, and vendors. It should make authority and evidence visible without turning every low-risk improvement into a centralized bottleneck.
Read the field guideEcosystem
How models and transaction protocols connect the wider market.
Agent commerce
Agent commerce separates discovery, offer details, checkout coordination, user authorization, payment execution, fulfillment, and receipts. Protocols can make these steps interoperable, but accountable consent and auditable state remain application responsibilities.
Read the field guideModel strategy
Model strategy matches workload requirements to model capability, latency, cost, context, tool use, and operational controls. A production choice should be based on representative evaluation and a migration path, not a permanent assumption that one provider or benchmark is universally best.
Read the field guide