Official
Read and write local files with configurable access controls.
filesystemfilesreadwrite
Builder Favorite MCPs
The MCPs that developers install first and return to most often. These show up in nearly every AI-builder stack we track — the Swiss-Army-knife set. If you are spinning up a new workstation, start here.
9 MCPs in this collectionFilter:
builder-favoriteTop Builder Favorites MCPs ranked
Popularity-ordered. Click any card for install snippets, fact sheet, and trust signals.
Local SQLite database access with full read/write support.
databasesqlitelocalstorage
Official
Local Git operations: commit, diff, log, branch, and more.
gitlocalcommitsdiff
Official
Full GitHub API access: repos, PRs, issues, and code search.
gitgithubreposprs
Structured step-by-step reasoning for complex problem solving.
reasoningthinkingagentsplanning
Persistent knowledge graph memory across AI conversations.
memorypersistenceknowledge-graphagents
Up-to-date library docs pulled directly into your AI context.
codingdocumentationlibrariesaccuracy
Full Supabase access: database, auth, storage, and edge functions.
supabasedatabasebackendpostgres
Manage Linear issues, projects, and cycles from AI context.
linearproject-managementissuesproductivity
FAQ: Builder Favorites
How did you pick the "builder favorites"?
Based on configs we see in the wild — public .mcp.json files, client-specific guides, and what shows up in the most frequently shared Claude/Cursor setups. It is a signal, not a formal survey, and it skews toward developer workflows.
Is this the same as the top-rated list?
No. "Top-rated" is a composite of scores across setup, flexibility, popularity. "Builder favorites" is narrower — it captures what engineers reach for daily, which is often a smaller, more opinionated set.
What should I add after the builder basics?
Whatever matches your stack: a database MCP (Postgres or Supabase) if you work with data, the GitHub MCP if you review PRs, Puppeteer if you test or scrape web UIs, and E2B if your agents execute code.