sem-mcp

Semantic version control built on Git.
Instead of lines changed, sem tells you what entities changed: functions, methods, classes.

sem is a semantic version control tool that works on top of Git. It parses your code with tree-sitter, extracts every function, class, and method as an entity, and diffs at the entity level instead of lines. This means you see "functionblahhwas modified" instead of "lines x-y changed."

curl -fsSL https://raw.githubusercontent.com/Ataraxy-Labs/sem/main/install.sh | sh

Or install the npm wrapper intonode_modules:

npm install --save-dev @ataraxy-labs/sem

With Bun, trust the package so itspostinstallscript can download the binary:

bun add -d @ataraxy-labs/sem bun pm trust @ataraxy-labs/sem

Once installed, update to the latest release any time:

cargo install --git https://github.com/Ataraxy-Labs/sem sem-cli

docker build -t sem . docker run --rm -it -u "$(id -u):$(id -g)" -v "$(pwd):/repo" sem diff

# Option 1: alias in your shell profile (~/.bashrc, ~/.zshrc) alias sem="$HOME/.cargo/bin/sem" # Option 2: make sure cargo bin comes first in PATH export PATH="$HOME/.cargo/bin:$PATH" # Option 3: if installed via Homebrew export PATH="$(brew --prefix)/bin:$PATH"

If you installed via npm/bun, the binary lives innode_modules/.bin/semand is invoked throughnpx semorbunx sem, which avoids the conflict entirely.

Works in any Git repo. No setup required. Also works outside Git for arbitrary file comparison.

sem stores its SQLite entity cache outside the repository, under the OS cache directory by default. SetSEM_CACHE_DIR=/path/to/cacheto override the cache root; repo-local overrides are ignored so cache files do not dirty the working tree.

Entity-level diff with rename detection, structural hashing, and word-level inline highlights.

Cloud acceleration (for scale and teams)

Local is always free and, aftersem setup, always warm — the resident graph keeps your repo hot on your own machine, so day-to-day queries are instant with no login. You do not pay to make your laptop fast.

Cloud is for what a laptop can't do. On a very large monorepo the first local graph build can take a few seconds; a shared team graph shouldn't be rebuilt per developer; and CI wants the graph without checking anything out.sem loginconnects those cases to sem cloud, which keeps a warm, pre-built graph for your registered repos and serves the heavy queries from it (on a large repo like deno, animpactquery is ~86ms from the cloud vs ~573ms rebuilt locally).

sem login # GitHub device flow, one time sem impact myFunc --file src/foo.rs # served from the cloud's warm graph

- Not logged in, or the cloud is unreachable? sem computes locally and prints the exact same output. No failures, no difference in results.
- SEM_LOCAL=1forces local computation even when logged in.
- Small repos see no change, local is already fast. The win is for large codebases where rebuilding the graph each time is the bottleneck.

32 programming languages with full entity extraction via tree-sitter:

Everything else falls back to chunk-based diffing.

Custom extensions and extensionless files

For files with non-standard extensions, create a.semrcin your project root:

sem also reads.gitattributespatterns (diff=andlinguist-language=) if you already have those set up..semrctakes priority when both define the same extension.

For files with no extension at all, sem detects the language automatically from content (imports, declarations, shebang lines, vim modelines). This covers 19 languages with no config needed.
- Exact ID match— same entity in before/after = modified or unchanged
- Structural hash match— same AST structure, different name = renamed or moved (ignores whitespace/comments)
- Fuzzy similarity— >80% token overlap = probable rename

This means sem detects renames and moves, not just additions and deletions. Structural hashing also distinguishes cosmetic changes (whitespace, formatting) from real logic changes.

Why an agent wants these: instead of reading whole files and burning tokens, it can ask "what breaks if I changesubmitOrder" (sem_impact) or "give me just the context to refactor this function" (sem_context) and get a precise answer from the dependency graph.

Add it once, then talk to your agent normally. It calls the tools on its own.

Or one command that also installs the skill, so the agent knowswhento reach for sem:

Cursor, Claude Desktop, or any client with anmcpServersconfig:

- tree-sitterfor code parsing (native Rust, not WASM)
- git2for Git operations
- rayonfor parallel file processing
- xxhashfor structural hashing
- Plugin system for adding new languages and formats

sem collects anonymous usage data: the command name (e.g.diff,impact), CLI version, and operating system. Nothing else — no code, file paths, repo names, or user identity. Events are batched locally and sent in the background, so commands never wait on the network.

export SEM_NO_TELEMETRY=1 # or DO_NOT_TRACK=1

Automate Google Jules, the AI coding assistant, for tasks like code reviews, repository management, and AI-powered development workflows.

Token-efficient MCP server for GitHub source code exploration via tree-sitter AST parsing

Code intelligence MCP server - PageRank, blast radius, co-change, hotspots, clone detection across 34 languages in a single Rust binary.

Refactor TypeScript and JavaScript codebases using ts-morph. Supports renaming symbols and files, moving symbols, and searching for references.

MCP server that gives AI coding agents real-time npm & PyPI package versions, metadata, and dependency info.

Integrates with the Gerrit code review system to review code changes and details.

Make git commits on behalf of AI to track AI contributions in your codebase.

Browse GitHub repositories, explore directories, and view file contents.