{"slug": "six-open-source-pieces-one-javascript-agent-stack", "title": "Six open-source pieces, one JavaScript agent stack", "summary": "The open-source AgentsKit ecosystem provides a set of independent JavaScript and TypeScript packages for building agent stacks, including runtime, memory, tools, evaluation, and UI components. Its core has zero runtime dependencies and a 10 KB gzipped budget, with six formal contracts ensuring substitutability. The ecosystem includes AgentsKit Chat for interactive agent experiences, a Registry for distributing ready-to-use agents, Agents Playbook for engineering patterns, Doc Bridge for documentation handoffs, and a Code Review CLI.", "body_md": "Most agent projects do not fail because the first model call is difficult. They become difficult when the prototype needs memory, tools, evaluation, a user interface, documentation that coding agents can navigate, and a repeatable review process.\n\nThat is the problem the open-source AgentsKit ecosystem is trying to solve for JavaScript teams. It is a set of independent projects with shared contracts, rather than one application that must own the whole stack.\n\n[AgentsKit](https://github.com/AgentsKit-io/agentskit) provides focused JavaScript and TypeScript packages for runtimes, adapters, tools, skills, memory, RAG, evaluation, observability, sandboxing, and UI bindings.\n\nIts core has zero runtime dependencies and a CI-enforced 10 KB gzipped budget. Six formal contracts keep adapters, tools, memory, retrievers, skills, and runtimes substitutable.\n\nYou can run a first local agent without an API key:\n\n```\nnpm install @agentskit/core @agentskit/runtime tsx\n```\n\nThen connect OpenAI, Anthropic, Gemini, Ollama, or another adapter without changing the rest of the runtime composition.\n\n[AgentsKit Chat](https://github.com/AgentsKit-io/agentskit-chat) defines an interactive agent experience once and renders it through React, React Native, Ink, Vue, Svelte, Solid, or Angular.\n\nIt also supports deterministic local answers before an optional backend call. That matters for documentation and support interfaces where exact project facts should not require an LLM.\n\n```\npnpm dlx @agentskit/chat-cli@0.4.0 init my-chat --renderer react --yes\n```\n\nThe [AgentsKit Registry](https://github.com/AgentsKit-io/agentskit-registry) distributes ready-to-use agents in a shadcn-style model: the CLI copies the source into your project, so you can inspect and modify it.\n\n```\nnpx agentskit add research\nnpx agentskit add code-review\n```\n\nThe public catalog currently includes research, code-review, and knowledge-promotion agents, and is open to contributions.\n\n[Agents Playbook](https://github.com/AgentsKit-io/agents-playbook) turns engineering practices for coding agents into inspectable patterns, prompts, templates, and zero-dependency gates.\n\nInstead of telling an agent to “write production-quality code,” a team can define package boundaries, required evidence, review rules, and executable checks. The repository currently verifies 87 production patterns, six engineering pillars, six SDLC phases, and 13 gate scripts from its own source.\n\n[Doc Bridge](https://github.com/AgentsKit-io/doc-bridge) converts repository documentation and ownership maps into deterministic handoffs:\n\n```\n{\n  \"startHere\": \"docs/for-agents/packages/auth.md\",\n  \"editRoots\": [\"packages/auth\"],\n  \"checks\": [\"pnpm --filter @demo/auth test\"],\n  \"humanDoc\": \"/docs/guides/auth\"\n}\n```\n\nThe same handoff is available through CLI, MCP, and CI. The core path requires no LLM or API key.\n\nTry the bundled proof:\n\n```\nnpx -y @agentskit/doc-bridge demo --text\n```\n\n[AgentsKit Code Review](https://github.com/AgentsKit-io/code-review-cli) runs locally or as a GitHub Action. It supports existing Codex and Claude CLI logins, hosted providers, gateways, and local Ollama models.\n\nSeven review lenses generate candidate findings. Independent verification votes then remove weak findings before the result reaches the developer.\n\n```\nnpx --yes github:AgentsKit-io/code-review-cli --provider codex-cli\n```\n\nThe output can be Markdown, GitHub review comments, or SARIF.\n\nYou do not need to adopt everything:\n\nEach repository is independently usable and openly licensed. The shared objective is interoperability: a team should be able to start with one package and add another without replacing its existing architecture.\n\nThe ecosystem is looking for more than implementation code. Useful contributions include provider adapters, framework examples, reusable agents, false-positive reductions, documentation adapters, executable gates, fixtures, and clearer onboarding.\n\nIf one of these problems matches work you are already doing, try the smallest relevant project, report what did not compose cleanly, and consider starring the repositories you want other JavaScript developers to discover.", "url": "https://wpnews.pro/news/six-open-source-pieces-one-javascript-agent-stack", "canonical_source": "https://dev.to/agentskit/six-open-source-pieces-one-javascript-agent-stack-2of4", "published_at": "2026-07-22 00:14:04+00:00", "updated_at": "2026-07-22 00:29:04.018903+00:00", "lang": "en", "topics": ["developer-tools", "ai-agents"], "entities": ["AgentsKit", "OpenAI", "Anthropic", "Gemini", "Ollama", "React", "Vue", "Svelte"], "alternates": {"html": "https://wpnews.pro/news/six-open-source-pieces-one-javascript-agent-stack", "markdown": "https://wpnews.pro/news/six-open-source-pieces-one-javascript-agent-stack.md", "text": "https://wpnews.pro/news/six-open-source-pieces-one-javascript-agent-stack.txt", "jsonld": "https://wpnews.pro/news/six-open-source-pieces-one-javascript-agent-stack.jsonld"}}