Command Palette

Search for a command to run...

UnylyUnyly
Весь каталог

Agentcast

БесплатноНе проверен

Structured-output enforcer: extract and validate JSON from messy LLM text.

GitHubEmbed

Описание

Structured-output enforcer: extract and validate JSON from messy LLM text.

README

npm tests mcp

An MCP server that gives AI assistants the ability to enforce structured output: extract JSON from messy LLM text, gate it against a shape spec, and produce the retry feedback message when the model returns the wrong shape.

Built on top of @mukundakatta/agentcast. Works with Claude Desktop, Cursor, Cline, Windsurf, Zed, and any other MCP client.

Tools exposed

extract_json

Pull a JSON value out of messy LLM output. Tries the whole text, then a fenced ```json ``` block, then the largest balanced {...} / [...] substring. Returns the parsed value plus which strategy succeeded.

{
  "text": "Sure, here you go:\n```json\n{\"answer\": 42}\n```\nLet me know!"
}

{
  "value": { "answer": 42 },
  "found": true,
  "source": "fenced_json"
}

source is one of whole, fenced_json, fenced_plain, balanced_substring, or none.

validate_response

Validate a parsed JSON value against an agentcast shape spec. Spec maps field name to type: string, number, boolean, array, object. Suffix with ? for optional.

{
  "value": { "name": "ada" },
  "shape": { "name": "string", "age": "number" }
}

{
  "valid": false,
  "error": "missing required field 'age'"
}

build_retry_prompt

Given an attempt history, produce the validation-error feedback message agentcast appends to the conversation when the model returned the wrong shape. Codifies the "validation error as feedback" pattern for non-Node MCP clients that want to drive the same retry loop manually.

{
  "attempts": [
    { "text": "{\"name\":\"ada\"}", "error": "missing required field 'age'" }
  ],
  "expected_shape": { "name": "string", "age": "number" }
}

{
  "feedback": "Your previous response did not match the required shape. Error: missing required field 'age'\n\nTry again. Respond with ONLY valid JSON that fixes the error above.\n\nExpected shape: {\"name\":\"string\",\"age\":\"number\"}"
}

Install

Claude Desktop

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "agentcast": {
      "command": "npx",
      "args": ["-y", "@mukundakatta/agentcast-mcp"]
    }
  }
}

Cursor / Cline / Windsurf / Zed

Same shape, in the appropriate mcp.json for your client. Most clients auto-discover via npx -y @mukundakatta/agentcast-mcp.

Local install

npm install -g @mukundakatta/agentcast-mcp
mcp-agentcast        # listens on stdio

Why this matters

When an LLM is supposed to return structured data, it sometimes wraps the JSON in prose, fences, or hallucinated fields. Standard JSON.parse throws. Hand-rolled regex misses nested structure. This MCP server gives any model driving an agent a real handle on (1) pulling JSON out of the response, (2) checking it matches the expected shape, and (3) building the exact retry prompt that nudges the model to fix it on the next turn.

License

MIT.

from github.com/MukundaKatta/agentcast-mcp

Установить Agentcast в Claude Desktop, Claude Code, Cursor

Рекомендуется · одна команда, все IDE
unyly install agentcast

Ставит в Claude Desktop, Claude Code, Cursor и VS Code — сам разбирается с npx, uvx и сборкой из исходников.

Впервые? Поставь CLI: curl -fsSL https://unyly.org/install | sh

Или настроить вручную

Выполни в терминале:

claude mcp add agentcast -- npx -y @mukundakatta/agentcast-mcp

FAQ

Agentcast MCP бесплатный?

Да, Agentcast MCP бесплатный — установка в пару кликов через Unyly без оплаты.

Нужен ли API-ключ для Agentcast?

Нет, Agentcast работает без API-ключей и переменных окружения.

Agentcast — hosted или self-hosted?

Self-hosted: сервер запускается локально на твоей машине командой из раздела установки.

Как установить Agentcast в Claude Desktop, Claude Code или Cursor?

Открой Agentcast на unyly.org, выбери вкладку своего клиента (Claude Desktop, Claude Code, Cursor) и нажми Install — конфиг сгенерируется автоматически, без правки JSON.

Похожие MCP

Compare Agentcast with

Не уверен что выбрать?

Найди свой стек за 60 секунд

Автор?

Embed-бейдж для README

Похожее

Все в категории ai