{
  "markdown": "# Elicitly — Free Edition\n\n[![npm — elicitly](https://img.shields.io/npm/v/elicitly?label=elicitly)](https://www.npmjs.com/package/elicitly)\n[![npm — @elicitly/tools](https://img.shields.io/npm/v/%40elicitly%2Ftools?label=%40elicitly%2Ftools)](https://www.npmjs.com/package/@elicitly/tools)\n[![Docs](https://img.shields.io/badge/docs-elicitly.ai-blue)](https://www.elicitly.ai/docs/)\n[![License: Apache-2.0](https://img.shields.io/badge/license-Apache--2.0-green)](./LICENSE)\n[![CI](https://github.com/elicitly/elicitly/actions/workflows/ci.yml/badge.svg)](https://github.com/elicitly/elicitly/actions/workflows/ci.yml)\n[![Coverage](https://codecov.io/gh/elicitly/elicitly/graph/badge.svg)](https://codecov.io/gh/elicitly/elicitly)\n[![elicitly MCP server — quality and maintenance score on Glama](https://glama.ai/mcp/servers/elicitly/elicitly/badges/score.svg)](https://glama.ai/mcp/servers/elicitly/elicitly)\n\n**Your AI has questions. Your users have answers.** Elicitly connects them —\nhuman-in-the-loop for prompts and [Agent Skills](https://agentskills.io/) over\n[MCP elicitation](https://modelcontextprotocol.io/docs/learn/client-concepts).\n\nThis repository is the open-source **Free Edition**: the `elicitly` local MCP\nserver and the `@elicitly/tools` library. The hosted **Pro Edition** — durable\napprovals with review pages, an audit trail, and organization dashboards —\nlives at [elicitly.ai](https://www.elicitly.ai).\n\n## Quick start\n\nRun the local (stdio) MCP server straight from npm — no clone, no config:\n\n    npx -y elicitly\n\nRegister it in your MCP host (Claude Code shown; any stdio host works):\n\n    claude mcp add elicitly -- npx -y elicitly\n\nOr skip the command line: in Claude Desktop, download `elicitly.mcpb` from\nthe [latest release](https://github.com/elicitly/elicitly/releases/latest)\nand open it — a one-click, self-contained\n[desktop extension](./packages/elicitly/README.md#claude-desktop-one-click).\n\nThen ask your agent to call `elicit_confirm` (\"Ship it?\"), `elicit_form`\n(your own JSON schema), or `elicit_doctor` (does this host actually support\nelicitation?).\n\n## Packages\n\n| Package | What it is |\n| --- | --- |\n| [`elicitly`](./packages/elicitly) | The local stdio MCP server (`npx -y elicitly`) |\n| [`@elicitly/tools`](./packages/tools) | Embeddable toolkit: register the elicitation tools on **your own** `McpServer` |\n\n## Why\n\nMCP has a native answer for asking a human — elicitation — but host support is\ninconsistent, and only an MCP *server* can trigger it. Nobody wants to roll\ntheir own server just to ask a question. Elicitly is that server (or, via\n`@elicitly/tools`, the toolkit inside yours), plus the diagnostic\n(`elicit_doctor`) that makes real-world host support visible — feeding the\npublic [Support Matrix](https://www.elicitly.ai/docs/elicitation/support-matrix/).\n\n## Develop\n\n    pnpm install\n    pnpm build        # tsdown builds both packages\n    pnpm test         # vitest\n    pnpm test:coverage # vitest + v8 coverage (thresholds enforced; HTML report in coverage/)\n    pnpm typecheck    # tsc\n    pnpm check        # biome\n\n## License\n\n[Apache-2.0](./LICENSE)\n",
  "bytes": 3124,
  "sha": "ab11dedd90fdf16ecc5bbf144403665561db13fded70d16debad4217ba3ac0b7",
  "repo_slug": "elicitly/elicitly",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_ai_elicitly_elicitly_99abac04/readme"
}