Meridian

#Why I built this

The investment team had a growing collection of AI tools — memo generation, network mapping, deal research, Slack search — but each lived in a separate app. Using them meant context-switching, copying data between windows, and manually stitching results together. I wanted a single surface where any Slack message could trigger any tool (or chain of tools) without leaving the conversation.

#What it does

Meridian is a TypeScript service that sits between Slack and multiple MCP (Model Context Protocol) servers. When someone sends a message, it parses intent, routes the request to the right server(s), executes multi-step tool chains, and streams results back into the originating thread. Adding a new capability is a single MCP server registration — no changes to the core router.

#Architecture

The system is built around three layers:

1. Slack ingress — Slack Bolt receives messages, extracts thread context (parent message, prior replies), and passes the full conversation to the intent parser
2. Intent routing — Claude API with tool-use definitions parses the message into one or more MCP tool calls. The router resolves which server owns each tool and batches calls that can run in parallel
3. MCP execution — each registered server exposes tools via the Model Context Protocol spec. Meridian maintains persistent connections to each server, handles retries, and multiplexes results back into a single Slack reply

#Multi-step chains

The interesting part is chaining. A message like "research this founder and draft a memo" triggers Atlas (network lookup) → Fathom (scoring) → Dossier (memo generation) as a sequential pipeline. Each step's output becomes context for the next. The chain definition lives in a simple JSON config — no orchestration code needed.

#Server registration

New tools register by pointing Meridian at their MCP endpoint:

{
  "name": "dossier",
  "endpoint": "https://dossier.internal/mcp",
  "tools": ["generate_memo", "research_founder", "research_market"]
}

The router discovers available tools at startup and re-discovers on a configurable interval. Hot-reloading means deploying a new tool server makes it available to Slack users within seconds.

#Thread context management

Slack threads are the UX primitive. Every tool invocation preserves thread context — if you ask a follow-up question, the system has the full history of what was already retrieved or generated. Context is stored in a lightweight KV store (Vercel KV) with a TTL matching Slack's thread activity window.

#Technical decisions

• MCP over custom RPC — MCP is an open standard with a growing ecosystem. Using it means any MCP-compatible server can plug into Meridian without adapter code. The protocol handles tool discovery, parameter schemas, and streaming natively.
• Claude tool-use for routing — intent classification via tool-use definitions is more robust than keyword matching or fine-tuned classifiers. Adding a new tool to the routing layer is just adding a tool definition to the prompt — no retraining.
• Slack Bolt over webhooks — Bolt handles socket mode, retry logic, and rate limiting out of the box. The alternative was managing webhook verification, challenge responses, and retry deduplication manually.

#Stack

TypeScript, Slack Bolt (socket mode), Claude API (tool-use), Model Context Protocol SDK, Vercel KV, Vercel (deployment)