Praxicraft Assess MCP
Enables agents to connect to the Praxicraft Assess Public API to invite candidates, manage assessments, pipelines, and webhooks, and fetch results via OAuth or API-key authentication.
README
Praxicraft Assess MCP
Official Model Context Protocol server for the Praxicraft Assess Public API.
Connect Cursor, Claude Code, Claude Desktop, Codex, or any MCP client so an agent can invite candidates, manage assessments and pipelines, register webhooks, and fetch results — with your permission.
npx -y @praxicraft/assess-mcp
Requires Node.js 18+. Product docs: docs.praxicraft.com/assess-mcp · Build with agents
Prefer one install for API MCP + docs search + skills? Use the Assess Agent Plugin.
Table of Contents
- Two ways to connect
- Authentication
- Quickstart
- What agents can do
- Live vs Test mode
- Errors
- Requirements & support
- License
Two ways to connect
| Mode | Best for | Auth |
|---|---|---|
| Hosted HTTP | Interactive IDE sessions | Browser OAuth at https://assess.praxicraft.com/mcp (always live) |
Stdio (npx @praxicraft/assess-mcp) |
CI, locked-down machines, Test mode | PRAXICRAFT_API_KEY (ct_live_… or ct_test_…) |
Hosted OAuth is the easiest path in Cursor and Claude. Use stdio when you need a ct_test_ key or a non-interactive environment.
Authentication
Hosted (OAuth)
No API key in your config. The first tool call opens a browser consent screen. Owners, admins, and developers can approve.
Hosted sessions are always live mode.
Stdio (API key)
Create an organisation API key:
Assess → Developer → API Keys → create key → copy ct_live_… or ct_test_… (shown once).
export PRAXICRAFT_API_KEY="ct_test_xxxxxxxxxxxxxxxx"
# optional
export PRAXICRAFT_API_BASE_URL="https://assess.praxicraft.com"
Never commit API keys. Prefer environment variables or a secrets manager.
IP allowlists on API keys apply to Public API / stdio MCP requests. They do not apply to hosted OAuth. Use a dedicated key with an empty allowlist for stdio if other keys are CIDR-restricted.
Scopes and rotation: Authentication
Recommended scopes
| Flow | Scopes |
|---|---|
| Invite + notify | assessments:read, invitations:write, candidates:read, webhooks:write |
| Pipelines | pipelines:read, pipelines:write, webhooks:write |
| Read-only | assessments:read, candidates:read, organisation:read |
Quickstart
Hosted HTTP (Cursor / Claude Code)
Add to your MCP settings:
{
"mcpServers": {
"praxicraft-assess": {
"url": "https://assess.praxicraft.com/mcp"
}
}
}
Claude Code example:
{
"mcpServers": {
"praxicraft-assess": {
"type": "http",
"url": "https://assess.praxicraft.com/mcp"
}
}
}
Restart the client, then try: “List my Assess assessments and invite me to the first active one.”
Stdio (API key)
{
"mcpServers": {
"praxicraft-assess": {
"command": "npx",
"args": ["-y", "@praxicraft/assess-mcp"],
"env": {
"PRAXICRAFT_API_KEY": "ct_test_xxxxxxxxxxxxxxxx"
}
}
}
}
From source (contributors):
git clone https://github.com/praxicraft-platform/praxicraft-assess-mcp.git
cd praxicraft-assess-mcp
npm ci
npm test
npm run smoke
What agents can do
Tools map to the Assess Public API (/api/v1/public/…). Coverage includes:
| Area | Examples |
|---|---|
| Assessments | List, create, update, activate, attach cases |
| Invitations | Invite, bulk invite, remind, cancel |
| Results | List / retrieve by invite_token |
| Webhooks | Create destinations, test, list deliveries, retry |
| Pipelines | List, enroll, hold / unhold / reject |
| Interviews | List / manage interview rooms |
| Organisation | Profile, quota, squads, audit (plan-gated) |
| Integrations | Connector status (no secrets in responses) |
Responses are flat JSON (same shape as the Public API — no { "data": … } wrapper).
Remind agents to respect plan limits and to branch on error.code.
Live vs Test mode
Live (ct_live_… / hosted OAuth) |
Test (ct_test_… via stdio) |
|
|---|---|---|
| Data | Production org | Isolated test data |
| Webhooks | Live destinations only | Test destinations only |
A test candidate.passed never hits a live Slack or HTTPS destination. See Live and Test mode.
Errors
Public API errors look like:
{ "error": { "code": "SOME_CODE", "message": "…", "details": {} } }
Always branch on error.code. Reference: Errors
Requirements & support
- Node.js 18+ (stdio package)
- Assess plan that includes API keys (see plan limits)
- Product docs: docs.praxicraft.com
- Agent Plugin (API + docs Knowledge MCP + skills): praxicraft-assess-agent-plugin
- Issues: GitHub Issues
- Email: support@praxicraft.com
Maintainers deploying the hosted /mcp service: see DEPLOYMENT.md.
License
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.