Skip to main content
Glama

πŸ”Œ AI MCP Toolkit

An MCP (Model Context Protocol) server that exposes AI capabilities β€” two of which call other running services in this portfolio over HTTP, and one self-contained β€” as standardized tools any MCP-compatible client can use directly, including Claude Desktop.

Python MCP Groq Tavily


🎯 What It Does

MCP standardizes how an AI client calls external code. Write a tool once as an MCP server, and any MCP-compatible client β€” Claude Desktop, Claude Code, Cursor β€” can discover and call it, with no client-specific integration work.

This server exposes three tools, two of which are thin clients calling other repos' running FastAPI services over HTTP β€” not reimplementations of similar logic:

web_research   β†’ HTTP call to ai-research-agent's /research/ endpoint (port 8003)
review_pr       β†’ HTTP call to ai-pr-reviewer's /pr-review/ endpoint (port 8004)
explain_concept  β†’ self-contained, calls Groq directly β€” genuinely new, no reuse claim

The distinction matters and is worth being precise about: web_research and review_pr have no logic of their own. If ai-research-agent or ai-pr-reviewer isn't running, those tools fail outright β€” they don't fall back to anything. That failure mode is the proof that this is real cross-service interconnection, not two repos that happen to do similar things.


Related MCP server: Friday MCP Server

πŸ“Έ Screenshots

Running server.py

MCP Inspector β€” tool schemas and live testing MCP Inspector Local dev tool showing all 3 tools auto-discovered from @mcp.tool() decorators, with their generated input/output schemas.

Connected and running in Claude Desktop Claude Desktop connection Settings β†’ Developer β†’ Local MCP servers, showing ai-toolkit with status running.

A real tool call inside a Claude conversation Tool call in action Claude recognizing a request matches the web_research tool, invoking it, and returning a result grounded in live search β€” not its own training data.

Running inter_server_communication.py

Without MCP Inspector β€” PR REVIEW MCP Inspector Claude made a normal call to view the public PR and based on diff gives the suggestion.

With MCP Inspector - PR REVIEW MCP Inspector Now claude made custom mcp tool calls and forwarded the request to ai-pr-reviewer over HTTP and returned the identical structured result.

Side by side, these two screenshots are the actual evidence of cross-service reuse β€” same backend, same output, two different ways of reaching it.


✨ Features

  • Protocol-standard tool exposure β€” built on the official MCP Python SDK (FastMCP)

  • Real cross-repo interconnection β€” web_research and review_pr are HTTP clients of other services in this portfolio, isolated in their own module for clarity

  • Auto-generated schemas β€” tool input/output schemas derive from Python type hints and docstrings

  • Honest dependency, not duplication β€” if a backing service is down, its tool fails; nothing is silently reimplemented as a fallback

  • Client-agnostic β€” works with Claude Desktop, Claude Code, MCP Inspector, or any future MCP client unchanged


πŸ—οΈ Architecture

Claude Desktop (or any MCP client)
        β”‚  stdio + MCP protocol
        β–Ό
server.py  (FastMCP β€” tool registration, schema generation)
        β”‚
        β”œβ”€β”€ explain_concept ──────────────► Groq directly (no other service)
        β”‚
        └── inter_server_communication.py
                β”œβ”€β”€ web_research ─── HTTP ──► ai-research-agent  (port 8003)
                └── review_pr     ─── HTTP ──► ai-pr-reviewer     (port 8004)

inter_server_communication.py is a separate module specifically because it carries the cross-service dependency β€” keeping it isolated from server.py makes the "this tool depends on another repo being up" relationship explicit and easy to point to, rather than buried inside tool definitions.


🧠 How It Works

server.py registers tools with @mcp.tool(). Two of those tools don't contain business logic β€” they import functions from inter_server_communication.py, which makes an HTTP POST to another repo's running FastAPI service and returns its response, reshaped into a readable string for the MCP client.

# inter_server_communication.py
def call_research_agent(topic: str, depth: str = "quick") -> str:
    response = httpx.post(
        f"{RESEARCH_AGENT_URL}/research/",
        json={"topic": topic, "depth": depth},
        timeout=120
    )
    response.raise_for_status()
    data = response.json()
    findings = "\n".join(f"- {f}" for f in data.get("key_findings", []))
    return f"{data.get('summary', '')}\n\nKey findings:\n{findings}"
# server.py
@mcp.tool()
def web_research(topic: str, depth: str = "quick") -> str:
    """Run the ai-research-agent service's autonomous web research agent on a topic."""
    return call_research_agent(topic, depth)

When Claude calls review_pr with a GitHub PR URL, the request goes: Claude β†’ MCP server β†’ inter_server_communication.py β†’ HTTP β†’ ai-pr-reviewer's /pr-review/ endpoint β†’ GitHub API (to fetch the diff) β†’ Groq (to generate the review) β†’ back through the same chain to Claude. Five hops, three repos, one conversational request.


πŸ—‚οΈ Project Structure

ai-mcp-toolkit/
β”œβ”€β”€ server.py                        # Tool registration, FastMCP entry point
β”œβ”€β”€ inter_server_communication.py    # HTTP clients for ai-research-agent and ai-pr-reviewer
β”œβ”€β”€ .env.example
└── .gitignore

πŸš€ Getting Started

Prerequisites

Installation

git clone https://github.com/vyavahare-kishor/ai-mcp-toolkit
cd ai-mcp-toolkit

curl -LsSf https://astral.sh/uv/install.sh | sh
uv venv
source .venv/bin/activate
uv add "mcp[cli]" httpx groq python-dotenv

Configuration

cp .env.example .env
GROQ_API_KEY=your_groq_api_key_here
RESEARCH_AGENT_URL=http://localhost:8003
PR_REVIEWER_URL=http://localhost:8004

Run the dependent services first

# Terminal 1 β€” ai-research-agent
cd ai-research-agent && uvicorn main:app --reload --port 8003

# Terminal 2 β€” ai-pr-reviewer
cd ai-pr-reviewer && uvicorn main:app --reload --port 8004

Test locally β€” MCP Inspector

uv run mcp dev server.py

# for cross server testing
uv run mcp dev inter_server_communication.py

Try review_pr with a real public GitHub PR URL while watching ai-pr-reviewer's terminal β€” you should see the incoming request logged there, confirming the call actually crossed into that repo.

Connect to Claude Desktop

{
  "mcpServers": {
    "ai-toolkit": {
      "command": "uv",
      "args": ["--directory", "/absolute/path/to/ai-mcp-toolkit", "run", "server.py"]
    }
  }
}

# for cross server testing
{
  "mcpServers": {
    "ai-toolkit": {
      "command": "uv",
      "args": ["--directory", "/absolute/path/to/ai-mcp-toolkit", "run", "inter_server_communication.py"]
    }
  }
}

Restart Claude Desktop, then ask: "Use review_pr to review this PR: [github PR url]"


πŸ—ΊοΈ Roadmap

  • Wrap ai-customer-support-bot's /support/ask as a 4th cross-service tool

  • Add a fallback message (not silent failure) when a dependent service is unreachable

  • Add an MCP resource exposing recent research/review history

  • Authentication for remote deployment beyond local stdio


Part of an AI-native engineering portfolio. Full journey: ai-engineering-journey

Project

Relationship to this one

ai-research-agent

web_research calls this repo's /research/ endpoint directly over HTTP β€” a real runtime dependency

ai-pr-reviewer

review_pr calls this repo's /pr-review/ endpoint directly over HTTP β€” same dependency relationship

ai-analyst-crew

Same Groq backend pattern, but no cross-service calls β€” useful contrast


πŸ‘¨β€πŸ’» Author

Kishor Vyavahare Senior Software Engineer β†’ AI Native Engineer

11+ years of backend engineering (Ruby on Rails, PostgreSQL, AWS). Now building production AI systems β€” RAG pipelines, agents, multi-agent crews, and protocol-standard tool exposure with real cross-service architecture.

LinkedIn GitHub


πŸ“„ License

MIT License β€” use it, fork it, build on it.

Available Tools

3 tools
explain_conceptA

Explain a technical or AI concept tailored to a specific audience's background level, using concrete analogies.

Inputs: concept (required), audience. Returns: result

review_code_diffA

Review a code diff and return structured, actionable feedback. Pass raw diff text. Optionally specify a focus area like 'security'.

Inputs: diff (required), focus. Returns: result

web_researchA

Search the web for current information on a topic and return a concise summary with key points drawn from the results.

Inputs: topic (required). Returns: result

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    A3.6/5.0

    Scored across 3 tools

    Disambiguation5/5

    Each tool targets a completely different functionβ€”concept explanation, code review, and web researchβ€”with no overlap in purpose or inputs. An agent would not confuse them.

    Naming Consistency3/5

    Two tools follow a clear verb_noun pattern (explain_concept, review_code_diff), but web_research reverses the order (noun_verb if 'research' is the verb) or could be read as noun_noun, breaking consistency.

    Tool Count3/5

    With only 3 tools, the server feels minimal for a general-purpose AI toolkit. While the count itself is acceptable, the scope is too narrow to be considered well-rounded.

    Completeness2/5

    The tools are a random collection with no domain coherence. Missing fundamental AI capabilities like text generation, summarization, or image analysis, making the surface severely incomplete for a toolkit.

    Maintenance

    ActivityInactive
    ResponsivenessNo recent community issues

    Related MCP Connectors

    Related MCP Servers