get_project_context
Returns a verified, compact project brief: stack, auth, hosting, key integrations, and entry points without reading the whole repo.
DESCRIPTION
Use this first on an unfamiliar repository. It gives an orientation layer that is grounded in code and config, not guesses, so an agent can start with the right mental model before drilling into files.
WHEN TO USE
GOOD FIT
- You just opened a new repo and need orientation.
- An AI agent needs project context before making changes.
- You want to know what auth, database, hosting, and billing systems are present.
- You need the key entry points and environment variables before deeper analysis.
AVOID IT WHEN
- You already know the project layout well.
- You need the implementation of a specific function or class; use read_code.
- You need repo-wide pattern matching; use find_code.
PARAMETERS
| Name | Type | Required | Description |
|---|---|---|---|
| path | string | Yes | Absolute or relative path to the repo root. |
EXAMPLE
tools-call.json
{ "jsonrpc": "2.0", "id": 1, "method": "tools/call", "params": { "name": "get_project_context", "arguments": { "path": "/workspace/repo" } }}OUTPUT
response.json
{ "jsonrpc": "2.0", "id": 1, "result": { "content": [ { "type": "text", "text": "Stack: Next.js 16, TypeScript, Supabase\nAuth: Bearer API keys + session auth\nHosting: Vercel + Render\nKey files: src/app/api/mcp/route.ts, src/lib/auth.ts\nIntegrations: Stripe, Upstash Redis, Supabase" } ] }}TOKEN EFFICIENCY
This tool is deliberately compact. It replaces a scattered read of config files, package manifests, and entrypoints with a short verified brief.