Skip to content

README.md

Latest commit

 

History

History
111 lines (80 loc) · 3.6 KB

README.md

File metadata and controls

111 lines (80 loc) · 3.6 KB
Raw

AIMLInterviews MCP Server

aimlinterviews-mcp turns MCP-compatible AI assistants into interview coaches backed by the local AIMLInterviews curriculum.

It discovers practice problems, provides progressive hints, reviews candidate answers, and builds role- or company-focused preparation plans. The tutor follows a no-spoilers policy: problem retrieval excludes reference answers, hints provide progressively deeper process guidance, and answer review never executes submitted code.

Requirements

  • Node.js 20 or newer
  • A local clone of the AIMLInterviews repository
  • An MCP-compatible client such as Claude Code or Codex

Quick start

Clone the repository first:

git clone https://github.com/alirezadir/AIMLInterviews.git
cd AIMLInterviews

Then connect the tutor from the repository directory.

Claude Code

claude mcp add aimlinterviews -- npx -y aimlinterviews-mcp

Codex

codex mcp add aimlinterviews -- npx -y aimlinterviews-mcp

The server searches its working directory and parent directories for the clone. For clients launched elsewhere, configure the repository path explicitly:

{
  "mcpServers": {
    "aimlinterviews": {
      "command": "npx",
      "args": ["-y", "aimlinterviews-mcp"],
      "env": {
        "AIMLINTERVIEWS_ROOT": "/path/to/your/AIMLInterviews/clone"
      }
    }
  }
}

AIMLINTERVIEWS_ROOT must point to the repository root that contains README.md and src/.

Tools

Tool Purpose
get_server_status Show the repository root, source commit, and catalog counts
list_problems Filter practice problems by area, difficulty, tag, company mention, or query
get_problem Retrieve a prompt and source metadata without its answer
get_hint Request spoiler-safe hints at levels 1, 2, and 3
search_curriculum Find relevant chapter sections and source locations
get_learning_path Build a role- and experience-based study sequence
get_company_prep Build a role-based company preparation plan
review_answer Review reasoning against an interview rubric without executing code

The server also provides the interview_tutor, mock_interview, and study_plan prompts, plus aimlinterviews://catalog and aimlinterviews://tutor-policy resources.

Content and safety model

  • The server reads public Markdown under src/; it does not bundle a duplicate question bank.
  • Stable catalog IDs are derived from the source path and title.
  • Every problem and section includes its source path, line, and Git commit.
  • Repository reads are path-confined, size-limited, and skip symbolic links.
  • Company plans are curriculum-based role recommendations, not claims about private or current company interview questions.
  • Candidate code is treated as text and is never executed.
  • MCP protocol messages use stdout; diagnostics use stderr.

Restart the MCP client after pulling curriculum updates so the server rebuilds its in-memory catalog.

Development

cd MCP
npm install
npm test
npm pack --dry-run

Run the compiled server directly:

AIMLINTERVIEWS_ROOT="$(cd .. && pwd)" npm start

Inspect it interactively:

npm run inspect

Publishing requires npm access to the aimlinterviews-mcp package:

npm login
npm publish --access public

Design reference

The tool categories and tutor workflow were informed by the TorchLeet MCP server. This implementation is original and reads the AIMLInterviews curriculum rather than copying TorchLeet content.