Skene MCP Server

Skene reads your codebase and produces a **user journey** for your product: the lifecycle stages a user moves through, the milestones inside each

Local serverstdioPython

What is the Skene MCP server?

If you already use Skene, the skene mcp server is the piece that lets your assistant work with it directly. Skene reads your codebase and produces a user journey for your product: the lifecycle stages a user moves through, the milestones inside each stage, and the code or database evidence that proves each milestone is real.

What the server does

  • Agentic analysis — — two parallel agents read your project: one walks the codebase, one walks your SQL schemas. They each propose candidate milestones, which are then merged and classified into a canonical seven-stage lifecycle
  • Evidence-backed milestones — — each milestone carries a tracked_event, a confidence score, and one or more evidence chips pointing back to the exact code path, database table, or config file that justifies it
  • Journey visualizer — — the TUI opens a local web app that lays out the stages as columns, with milestone cards and clickable evidence chips, so you can review the journey alongside the underlying source
  • Code-only, schema-only, or both — — run against just a codebase, just a directory of *.sql files, or both for richer evidence

Installation

The server ships on PyPI as skene, so your MCP client can launch it on demand — there is no separate build step. Add the server block to your client's configuration, restart it, and the tools register themselves.

Available tools

The toolset is worth reading before you wire it up, because it tells you what the integration is really for:

  • Seven — stage user journey** — every product maps to the same canonical stages (*discovery, onboarding, activation, engagement, retention, expansion
  • Evidence — backed milestones** — each milestone carries a tracked_event, a confidence score, and one or more evidence chips pointing back to the exact code
  • Code — only, schema-only, or both** — run against just a codebase, just a directory of *.sql files, or both for richer evidence
  • Default — ** ./skene-context/journey.yaml — the analysed user journey for your product
  • Override — ** set output_dir in .skene.config or pass -o/--output to write elsewhere
  • Directory — Description

Worth knowing first

  • It runs with your machine's permissions. That is convenient and also the reason to think about what you point it at before you approve a tool call.
  • Keep per-call confirmation enabled while you learn its behaviour; it is the cheapest safeguard you have.

Where it fits

This sits in the database access group, where several servers overlap in what they claim to do but differ sharply once you actually set them up. Skene's toolset — Seven, Evidence, Code and 3 more — is a fair guide to whether it matches your workflow. It is maintained by skenetechnologies; worth a glance at recent repository activity before you build anything load-bearing on it.

This entry was verified against Skene's own documentation before publication; SyncDev keeps the directory reviewed rather than auto-generated.

Available tools

ToolWhat it does
Sevenstage user journey** — every product maps to the same canonical stages (*discovery, onboarding, activation, engagement, retention, expansion, virality*) so journeys are comparable across products. Stage names are special
Evidencebacked milestones** — each milestone carries a tracked_event, a confidence score, and one or more evidence chips pointing back to the exact code path, database table, or config file that justifies it.
Codeonly, schema-only, or both** — run against just a codebase, just a directory of *.sql files, or both for richer evidence.
Default** ./skene-context/journey.yaml — the analysed user journey for your product.
Override** set output_dir in .skene.config or pass -o/--output to write elsewhere.
DirectoryDescription

How to install the Skene MCP server

{
  "mcpServers": {
    "skene": {
      "command": "uvx",
      "args": ["skene"]
    }
  }
}

Add to claude_desktop_config.json, then restart Claude Desktop.

Example prompts to try

  • Use Skene to Seven.
  • Use Skene to Evidence.
  • Use Skene to Code.

Frequently asked questions

It connects Skene to MCP-compatible AI assistants such as Claude and Cursor, exposing 6 tools (Seven, Evidence, Code, and more) that the assistant can call on your behalf. Instead of copying data back and forth by hand, the assistant works with Skene directly.