{
  "markdown": "# streamparse-mcp\n\n[![npm](https://img.shields.io/npm/v/@mukundakatta/streamparse-mcp.svg)](https://www.npmjs.com/package/@mukundakatta/streamparse-mcp)\n[![tests](https://img.shields.io/badge/tests-4%20passing-brightgreen.svg)](#)\n[![mcp](https://img.shields.io/badge/protocol-MCP-blue.svg)](https://modelcontextprotocol.io)\n\nAn [MCP](https://modelcontextprotocol.io) server that gives AI assistants the\nability to parse partial / messy / truncated JSON.\n\nBuilt on top of [`@mukundakatta/streamparse`](https://github.com/MukundaKatta/streamparse).\nWorks with Claude Desktop, Cursor, Cline, Windsurf, Zed, and any other MCP client.\n\n## Tools exposed\n\n### `parse_partial_json`\n\nRecover a JSON value from a possibly-truncated string. Always returns a valid\nvalue with synthetic closure of any open strings, arrays, or objects.\n\n```json\n{\n  \"text\": \"{\\\"type\\\":\\\"tool_use\\\",\\\"name\\\":\\\"edit_file\\\",\\\"input\\\":{\\\"path\\\":\\\"a/b.ts\\\",\\\"cont\"\n}\n```\n\n→\n\n```json\n{\n  \"value\": {\n    \"type\": \"tool_use\",\n    \"name\": \"edit_file\",\n    \"input\": { \"path\": \"a/b.ts\", \"cont\": null }\n  },\n  \"complete\": false,\n  \"path\": [\"input\", \"cont\"],\n  \"bytes_consumed\": 67,\n  \"confidence\": 0.65\n}\n```\n\n### `extract_json_from_text`\n\nStrip prose, ` ```json ` fences, and comments around a JSON value embedded in\nLLM output. Returns the first parseable value.\n\n```\nSure, here you go:\n```json\n{ \"answer\": 42 }\n```\nLet me know!\n```\n\n→ `{ \"answer\": 42 }`\n\n### `validate_json`\n\nStrict-mode RFC 8259 validator. Returns `ok=true` and the parsed value on\nsuccess, or `ok=false` with a precise byte position and error message on\nfailure.\n\n## Install\n\n### Claude Desktop\n\nAdd to `claude_desktop_config.json`:\n\n```json\n{\n  \"mcpServers\": {\n    \"streamparse\": {\n      \"command\": \"npx\",\n      \"args\": [\"-y\", \"@mukundakatta/streamparse-mcp\"]\n    }\n  }\n}\n```\n\n### Cursor / Cline / Windsurf / Zed\n\nSame shape, in the appropriate `mcp.json` for your client. Most clients\nauto-discover via `npx -y @mukundakatta/streamparse-mcp`.\n\n### Local install\n\n```bash\nnpm install -g @mukundakatta/streamparse-mcp\nmcp-streamparse        # listens on stdio\n```\n\n## Why this matters\n\nWhen an LLM is mid-tool-call and you need the assistant to reason about the\nhalf-formed JSON it just wrote, no other tool gives a usable answer. Standard\n`JSON.parse` throws. Regex extraction misses nested structure. This MCP server\ngives Claude (or whichever model is driving) a real handle on partial JSON,\nright where it lives.\n\n## License\n\nMIT.\n",
  "bytes": 2465,
  "sha": "c6fa56d9d3c1afb6c2e495e2fe403ef522ffae6846668afcf3015c76bea021ff",
  "repo_slug": "mukundakatta/streamparse-mcp",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_io_github_mukundakatta_streamparse_1d1dae20/readme"
}