Key facts

Privacy
Runs in your browser; no inputs leave the page.
Cost
Free and offline after load.
Best use
Early architecture decisions before framework selection.
Boundary
Decision aid, not procurement or security approval.

Choose the right tool

Free toolAgent Readiness Checker

Answer structured questions about determinism, tools, side effects, and review needs to decide whether to build a prompt, workflow, single agent, or multi-agent system.

Free toolAgent Spec Generator

Generate a downloadable markdown AI agent spec with a system prompt skeleton, tool definitions outline, guardrails, and evaluation checklist.

Free toolAgent Framework Chooser

Rank custom orchestration, LangGraph, CrewAI, OpenAI Agents SDK, and Claude Agent SDK against your language, hosting, graph, memory, eval, and team constraints.

Use them in order

Start with the readiness checker. If the verdict says prompt or workflow, use that answer as the design constraint and avoid agent theater. If the verdict says agent, generate a written spec before picking a framework. Only after the spec names tools, state, permissions, guardrails, and evals should you run the framework chooser.

The tools are intentionally conservative. They bias toward the least autonomous design that can do the job, because every extra tool and every extra autonomous step expands the test surface.

Sources used

  1. Agents SDKOpenAI Developer Docs. Accessed 2026-07-13.

    Defines agents as applications that plan, call tools, collaborate, and keep state.

  2. Building Effective AgentsAnthropic Engineering. Accessed 2026-07-13.

    A practical distinction between augmented LLMs, workflows, and autonomous agents.

  3. LangGraph OverviewLangChain Docs. Accessed 2026-07-13.

    Positions LangGraph as an orchestration runtime for durable execution, streaming, HITL, and persistence.

  4. CrewAI DocumentationCrewAI. Accessed 2026-07-13.

    Introduces CrewAI agents, crews, flows, guardrails, memory, knowledge, and observability.

  5. OpenAI Agents SDKOpenAI. Accessed 2026-07-13.

    Documents the Python Agents SDK primitives: agents, handoffs, guardrails, tracing, sessions, tools, and sandbox agents.

  6. Agent SDK OverviewClaude Code Docs. Accessed 2026-07-13.

    Documents Claude Agent SDK built-in tools, hooks, subagents, MCP, permissions, sessions, and production use cases.