Agentcast
by MukundaKattacommunityv0.1.0
1
An MCP server for extracting JSON from messy LLM output, validating parsed JSON against a simple shape spec, and generating retry feedback prompts when the shape is wrong. It is designed to work with MCP clients like Claude Desktop, Cursor, Cline, Windsurf, and Zed.
From the registry: Structured-output enforcer: extract and validate JSON from messy LLM text.
Install & configure
Prerequisites
- optional
Install in your client
Paste into the AI you're chatting with — it will figure out the rest
Please install the `agentcast` MCP server into my current AI client (that's you).
Optional prerequisites:
- Node.js and npm — Install Node.js and npm so the MCP server can be run via npx. (https://nodejs.org/'}],)
Canonical MCP server config (stdio transport):
- command: `npx`
- args: ["-y","@mukundakatta/agentcast-mcp"]
Note: README provides a portable stdio install via npx. A global npm install is also documented, but npx is the simpler preferred path. No environment variables are described.
Add this MCP server to my current client's config in the correct format for you. If you need secrets or credentials I haven't provided, ASK me — do not invent values or leave raw placeholders. After adding it, tell me how to verify the server is connected.Note: README provides a portable stdio install via npx. A global npm install is also documented, but npx is the simpler preferred path. No environment variables are described.
Tools (3)
extract_json
Pull a JSON value out of messy LLM output and report which extraction strategy succeeded.
validate_response
Validate a parsed JSON value against a shape spec with required and optional fields.
build_retry_prompt
Generate the validation-error feedback message to ask the model to retry with the correct JSON shape.
Packages
@mukundakatta/agentcast-mcp
npmv0.1.0transport: stdio
Details
- Language
- JavaScript
- License
- MIT
- Released
- Apr 27, 2026
- Last commit
- Jul 4, 2026
- Registry name
- io.github.MukundaKatta/agentcast
Categories
Related Servers
Claude Flow
AI orchestration with hive-mind swarms, neural networks, and 87 MCP tools for enterprise dev.
65.2k
Codebase Memory Mcp
Persistent codebase knowledge graph. Survives session restarts and context compaction.
33.0k
Serena
A powerful toolkit for coding, providing semantic retrieval and editing capabilities.
26.6k