Authentication
The server uses OAuth 2.0 — no API keys to copy or store. The first time you connect from any client:- Your browser opens to OnePerfectSlice
- You log in (or you’re already signed in)
- You select the team whose data this connection should access
- The client receives a token and reconnects automatically
Set up your client
Setup takes under a minute — add the URL, then authenticate. Pick your client:Claude Code
One CLI command. Best for developers in the terminal.
Claude Desktop
Add a custom connector. Best for non-technical users.
ChatGPT
Add a connector in settings. No terminal required.
Cursor
Add the URL in MCP settings.
VS Code
Connect through GitHub Copilot’s MCP support.
Codex
Configure and run
codex mcp login.Any other MCP client
Windsurf, Cline, Goose, Zed, LM Studio and more — the same URL works anywhere.
How it works
The MCP server sits on top of the OnePerfectSlice REST API and exposes 9 tools designed for agent interaction. Key differences from calling the API directly:- Composite operations —
run_slicestarts a run, polls for completion, and fetches evidence in one call, not three. - Flat parameters — no nested objects. Every filter is a top-level argument the agent can set directly.
- Guided errors — when something goes wrong, error messages tell the agent exactly which tool to call next.
Available tools
| Tool | What it does |
|---|---|
list_slices | Discover available analyses or get details for a specific slice |
preview_slice_count | Count matched calls before running (free, no AI cost) |
run_slice | Execute an analysis and get results + evidence |
get_run | Retrieve results from a past or in-progress run |
list_runs | Browse run history with optional filtering |
search_posts | Find summaries and scorecards matching your criteria |
get_post | Get a single post with full structured content |
search_evidence | Search evidence quotes across all completed runs |
get_filter_options | Discover valid filter values (call owners, companies, CRM data) |