algofox
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@algofoxContinue my last coding problem"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
AlgoFox MCP — coding interview practice in your terminal
Practice coding interviews with the coding agent you already use — Claude Code, Cursor, Codex, or any MCP client.
Works fully offline. Nothing ever leaves your machine. No account, no telemetry, no network calls — problems ship in the package and your progress lives in ~/.algofox/.
How it works
Say "Continue AlgoFox" (or pick a problem). The agent materializes it into
./algofox/<slug>/: the problem statement, a starter file, and a test runner.You write the solution — the agent is your interviewer/coach, not your ghostwriter. It clarifies, runs the tests when you ask, and fetches tiered hints.
Tests run locally (
python3 run_tests.pyornode run_tests.mjs). Results are recorded; solving earns XP and keeps your streak alive.
Close your terminal any time. Come back in any supported agent, say "Continue AlgoFox", and you resume where you left off.
Related MCP server: Engineering Knowledge Graph MCP Server
Install
Claude Code (recommended: the plugin)
The algofox-skill plugin bundles this server plus an interview-coach skill and /algofox commands:
/plugin marketplace add phuongnd11/algofox-skillOr add just the server:
claude mcp add algofox -- npx -y -p @algofoxdotapp/mcp algofox-mcpCursor
.cursor/mcp.json:
{ "mcpServers": { "algofox": { "command": "npx", "args": ["-y", "-p", "@algofoxdotapp/mcp", "algofox-mcp"] } } }Codex
~/.codex/config.toml:
[mcp_servers.algofox]
command = "npx"
args = ["-y", "-p", "@algofoxdotapp/mcp", "algofox-mcp"]For clients without skills, copy the coach block from AGENTS.md into your project's AGENTS.md / rules file so the agent behaves like an interviewer instead of solving problems for you.
Standalone CLI
Progress without opening an agent:
npx -p @algofoxdotapp/mcp algofox status # resume card
npx -p @algofoxdotapp/mcp algofox progress # full problem list
npx -p @algofoxdotapp/mcp algofox history # recent activity
npx -p @algofoxdotapp/mcp algofox resume # ready-to-paste agent promptWhat's inside
10 original coding problems (arrays & hashing to start — more topics coming) with typed test cases, tiered hints, and reference solutions.
5 languages: Python, JavaScript, TypeScript, Java, Go. You pick on first run; switch any time by telling your agent. LeetCode-style
Solutionboilerplate, sample input/output, and a test harness are generated from each problem's typed signature.Solutions are earned:
reveal_solutionunlocks after 2 failed runs or 2 hints. All content is local, so you can peek at the answer keys — but you'd only be cheating yourself.A run token ties submitted results to an actual test run, so your agent can't hand-wave a pass.
Want a full curriculum — guided lessons, spaced review, quick-play drills, company tracks? That's the AlgoFox app.
Requirements
Node ≥ 18. Per language: Python needs python3; TypeScript needs Node ≥ 22.6; Java needs a JDK (javac); Go needs the Go toolchain. JavaScript needs nothing extra.
License
Code: MIT. Problem content (content/): CC BY-NC-ND 4.0 — see content/LICENSE.
This server cannot be deployed
Maintenance
Related MCP Connectors
Remote MCP learning coach for coding agents.
Persistent memory and cross-session learning for AI coding assistants (hosted remote MCP).
Hosted MCP memory for coding agents: persistent across sessions, editable markdown, team sharing.
GodPrompt MCP + Agent Skill for coding agents with TDD, debugging, verification, and task routing.
Related MCP Servers
- AlicenseNot gradedqualityAmaintenanceA local-first study workspace for MCP agents, providing a web UI for learning, quizzing, and review, independent of agent lifecycle.4GPL 3.0
- AlicenseNot gradedqualityBmaintenanceProvides a stdio MCP bridge for coding agents to query and record engineering knowledge locally, preserving debugging history, failed attempts, and verified solutions.2 npmMIT
- FlicenseAqualityBmaintenanceEnables coding agents to discover unsolved Codeforces practice problems, analyze tag performance, and inspect submissions, profiles, rating history, and upcoming contests through MCP-compatible clients.6-
- FlicenseNot gradedqualityBmaintenanceEnables reproducible evaluation of AI coding agents by exposing repository inspection, code editing, test running, and deterministic verification through MCP tools.-