Show HN: Lain, a structural code graph and agent coordinator for coding agents

LAIN, a new structural code intelligence and coordination engine for AI coding agents, launched on Hacker News' Show HN, indexing codebases into an in-memory typed property graph via Tree-sitter, language servers, and Git commit history. LAIN exposes deterministic tools over the Model Context Protocol (MCP) — including get_blast_radius, get_call_chain, trace_dependency, and get_coupling_radar — to agents such as Claude Code, Cursor, Copilot, Codex, Agy, Cline, and Windsurf, and serves them over stdio or HTTP on port 9999 using JSON-RPC and SSE. The engine persists its graph to .lain/graph.bin with UUID v5 identifiers and tracks agent sessions and file claims through advisory presence registries to prevent overlapping edits.

Structural Code Intelligence & Multi-Agent Coordination for AI Assistants. Give your coding agents an in-memory graph brain instead of making them guess from flat text. LAIN is a persistent, high-performance code intelligence and coordination engine built specifically for AI coding agents Claude Code, Cursor, Copilot, Codex, Agy, Cline, Windsurf, etc. over the Model Context Protocol MCP . Instead of treating code as unstructured flat text or relying on fuzzy keyword searches, LAIN indexes your codebase into an in-memory, typed structural property graph using Tree-sitter, language servers, and Git commit history . It then exposes a rich suite of deterministic MCP tools that allow AI agents to navigate, reason about, and modify complex codebases without hallucinations, blind edits, or context-window waste. - Instant Answers : Federation overview, repo health, and call graphs answered in milliseconds. - Hot Reload : Edit repos.yaml or workspaces.yaml ; the server updates live without dropping a single active MCP session. - Interactive Tool Console : Exercise any MCP tool directly from the web browser; Copy as cURL gives agents and operators instant reproducibility. Note The demo is kept as a single GIF so it plays inline without storing duplicate video encodings in the repository. flowchart TB subgraph Clients "AI Agents & Developers" A "AI Coding Agent<br/ Claude Code / Cursor / Agy / Codex " B "Developer / Operator<br/ Browser Command Center " end subgraph Transports "MCP & Web Transports" S "stdio single-repo " H "HTTP :9999 JSON-RPC & SSE " end subgraph Core "LAIN Core Engine" EX "Unified MCP Tool Dispatcher" G "In-Memory Graph Engine<br/ Petgraph · UUID v5 " PRES "Presence & Claim Registry<br/ Advisory Leases & Locks " FED "Federation Engine<br/ N Repositories " end subgraph Sources "Code Analysis & Storage" TS "Tree-sitter AST Parser" LSP "Language Servers rust-analyzer, pylsp... " GIT "Git Commit History Co-change radar " BIN ".lain/graph.bin Persistent Cache " end A -- |MCP JSON-RPC| S A -- |MCP HTTP| H B -- |GET /| H S -- EX H -- EX EX -- G EX -- PRES EX -- FED G <-- BIN G -- TS G -- LSP G -- GIT 1. Indexing & Parsing : LAIN scans your code using Tree-sitter and language servers LSPs , extracting functions, classes, imports, and references into a property graph. 2. Persistent Graph Store : The graph is serialized into .lain/graph.bin using deterministically derived UUID v5 identifiers for instant reloads. 3. Temporal Mining : LAIN analyzes git commit logs to build a co-change coupling radar identifying modules that evolve together even without explicit imports . 4. Advisory Presence : In-memory and on-disk occupancy registries track agent sessions and file claims, preventing overlapping edits in real time. 5. Universal MCP Delivery : Exposes standardized tools over stdio or HTTP so any MCP-compatible agent can query the graph directly. LAIN provides specialized MCP tools categorized by capability: - get blast radius — Downstream impact analysis: every function, type, and file affected by changing a symbol. - get call chain — Shortest path between two functions in the call graph. - trace dependency — All upstream dependencies callees, imports, types of a target symbol. - get coupling radar — Files that frequently change together based on Git commit co-occurrence. - find anchors — Identifies the core architectural pillars most-called, most-stable symbols . - list entry points — Discovers main , HTTP routes, and event handlers. - get context depth — Measures abstraction distance from public entry points. - explore architecture — High-level hierarchical module and package tree. - register agent / heartbeat — Registers an agent session and keeps advisory leases fresh. - claim files / release files — Claims or releases files and symbol ranges before editing. - detect overlap — Analyzes overlapping symbol changes between git branches or concurrent sessions. - list active agents / who am i — Discovers other active agents and reports session identity. - semantic search requires ONNX model — see Setting Up Semantic Search setting-up-semantic-search-optional — Concept-based code search using local ONNX embeddings with hybrid BM25/stemmed ranking. - query graph — Composable JSON ops pipeline find , connect , filter , semantic filter , sort , limit . - explain symbol — Complete structural dossier for a symbol signature, callers, docstring, location . - list repos / get repo info — Status, health, and size of all repos registered in repos.yaml . - get federation health — Aggregate health counts, total node/edge counts, and a rough memory estimate across the federation. - get cross repo blast radius — Cross-repository impact analysis when modifying a shared symbol. - get cross repo blast radius for repo — Same as get cross repo blast radius , but the caller disambiguates the target repo by repo id instead of by symbol resolution. - search org — Organization-wide symbol and code search across all registered repositories. - find dead code — Detects unreachable functions and unused symbols excluding traits and tests . - suggest refactor targets — Identifies brittle code high-coupling, low-stability candidates . - get agent strategy — Retrieves operational guidelines and strategic instructions for agents. - get world state — Summarizes active sessions, file claims, and graph freshness in a single compact call. Install interactive — adds lain to PATH curl -fsSL https://raw.githubusercontent.com/spuentesp/lain/main/install.sh | bash Reload your shell, then verify source ~/.zshrc or ~/.bashrc lain --version See QUICKSTART.md https://github.com/spuentesp/lain/blob/main/docs/QUICKSTART.md for Homebrew, manual builds, non-interactive flags, and ONNX model setups. claude mcp add lain -- lain mcp lain setup --agent cursor Writes ~/.cursor/mcp.json the file Cursor reads directly . Windsurf shares the same config schema, so this command works for Windsurf too. lain setup --agent vscode Writes .vscode/mcp.json if it exists project-scoped or mcp.json in your user-config dir otherwise. Details in docs/COOKBOOK.md . lain setup --agent codex uses codex mcp add when the CLI is on PATH lain setup --agent continue writes ~/.continue/config.json lain setup --agent generic Writes .mcp.json at the workspace root. See docs/COOKBOOK.md for the exact schema. Run LAIN as a shared service across multiple repositories: lain server --config ./repos.yaml --transport http --port 9999 Access the Command Center UI in your browser at http://localhost:9999 . When lain server runs with --transport http , it serves the Command Center dashboard at GET / . It is a self-contained single-page application SPA that talks back to the running server over the same MCP JSON-RPC protocol. - Overview — Real-time node/edge stats, memory footprint, and federation health. - Graph — Interactive D3 force-directed visualizer of workspaces and symbol dependencies. - Repos — Repository table showing health, path, and node/edge statistics. - Query — Interactive query runner for query graph traversals. - Tools — Form-based MCP tool runner with auto-generated Copy as cURL snippets for quick testing. LAIN exposes the following CLI commands: | Command | Purpose | |---|---| | lain server | Start the MCP server the headline . Reads repos.yaml , serves MCP tools + the Command Center dashboard. Hot-reloads the config when it changes. | | lain mcp | Single-repo MCP server on stdio. Walks up from cwd for .git — the stable "drop in a clone and run" entrypoint. No repos.yaml required. | | lain setup | Guided onboarding: detects the repository and languages, optionally installs the semantic model, configures one MCP client --agent claude-code shells to claude mcp add ; --agent generic writes .mcp.json , and verifies the result with a real MCP round trip. --dry-run and --print-config change nothing. | | lain workspaces | Manage workspaces.yaml . Create, list, show, activate use , forget named groups of repos. | | lain repos | Manage repos.yaml . Add, list, remove a repo entry. | | lain query | Run a query graph ops-array against the project's persisted graph. | | lain oneshot | One-shot MCP query: boots a transient lain mcp server, sends a single tools/call , prints the result as a table, and exits. For "just grep the symbols without keeping a server alive". | | lain init | Scaffold a repos.yaml for the current directory. Walks up for .git , then writes a minimal config pointing at the discovered workspace. | | lain ask | Single-user LLM-assisted query uses semantic search when an embedding model is loaded; falls back to lexical heuristics via explain symbol . | | lain hooks | Agent pre-edit hook entry point: claim / release files, overlap-check for commit-time symbol overlap, lock / unlock for the zero-daemon filesystem-fallback layer. | | lain doctor | Read-only repository diagnosis. Reports binary identity, persisted graph freshness, structural and optional semantic capability states, installation paths, and an MCP initialize/tools-list probe. Use --json for the versioned machine-readable report and --workspace PATH outside the target clone. Exit codes are 0 ready, 1 usable but degraded, and 2 unusable. | | lain capabilities | Print the four canonical capability states and repository freshness. Add --json for agent-readable output. | | lain status | Print aggregate repository, index, and MCP readiness. Add --json for the versioned status object. | | lain schema | Emit the canonical tool-surface schema dump dump --out PATH defaults to ./docs/tool-schema.json . Pair with make schema && git diff --exit-code docs/tool-schema.json in CI to fail on schema drift. | | scripts/demo.sh | Capability demonstration and benchmark. Boots a real server against a synthetic repo whose call graph is known by construction, checks lain's answers against that ground truth not merely that it answered , then benchmarks the same tools against this repo at ~3.5k nodes. --quick skips the build and benchmark phases; --json FILE writes machine-readable results; --force-build overrides --quick / --no-build ; --allow-stale skips the binary-freshness check. Exits non-zero if any check fails or if the binary is older than any source file and --allow-stale was not passed . | lain server monitors repos.yaml and workspaces.yaml and automatically updates its federation state when they change—without restarting the process or dropping active agent connections. Both manual file edits and CLI commands lain repos add ... communicate over a local Unix socket ~/.local/lain/run/<repos-stem .sock to rebuild the index diff atomically. See docs/hot-reload.md https://github.com/spuentesp/lain/blob/main/docs/hot-reload.md for full operational details. | Doc | What's in it | |---|---| | docs/QUICKSTART.md https://github.com/spuentesp/lain/blob/main/docs/QUICKSTART.md | Step-by-step 5-minute setup and quickstart | | docs/USER MANUAL.md https://github.com/spuentesp/lain/blob/main/docs/USER MANUAL.md | Comprehensive operator and agent manual | | docs/ARCHITECTURE.md https://github.com/spuentesp/lain/blob/main/docs/ARCHITECTURE.md | System design, graph layering, and technical rationale | | docs/TECHNICAL.md https://github.com/spuentesp/lain/blob/main/docs/TECHNICAL.md | Deep dive into internal modules and data structures | | docs/FEDERATION.md https://github.com/spuentesp/lain/blob/main/docs/FEDERATION.md | Multi-repo setup, org search, and cross-repo intelligence | | docs/REPOS YAML.md https://github.com/spuentesp/lain/blob/main/docs/REPOS YAML.md | Schema and options for repos.yaml | | docs/multiplayer.md https://github.com/spuentesp/lain/blob/main/docs/multiplayer.md | Multi-agent coordination, claims, and advisory locks | | docs/hooks.md https://github.com/spuentesp/lain/blob/main/docs/hooks.md | Git and editor pre-edit hook configuration | | docs/query-language.md https://github.com/spuentesp/lain/blob/main/docs/query-language.md | Syntax and operations reference for query graph | | docs/quickstart-tools.md https://github.com/spuentesp/lain/blob/main/docs/quickstart-tools.md | Complete reference guide for all MCP tools | | docs/command-center.md https://github.com/spuentesp/lain/blob/main/docs/command-center.md | Features and usage of the web Command Center | | docs/INDEX.md https://github.com/spuentesp/lain/blob/main/docs/INDEX.md | Full documentation sitemap | | Requirement | Details | |---|---| | Rust build only | 1.75 or newer | | Git | Required for co-change coupling analysis | | ONNX Model | Optional — required only for semantic search | For semantic search , download an ONNX embedding model or run install.sh --download-model : mkdir -p .lain/models BAAI/bge-small-en-v1.5 recommended — 384d, ~120MB curl -L https://huggingface.co/BAAI/bge-small-en-v1.5/resolve/main/onnx/model.onnx \ -o .lain/models/model.onnx curl -L https://huggingface.co/BAAI/bge-small-en-v1.5/resolve/main/tokenizer.json \ -o .lain/models/tokenizer.json export LAIN EMBEDDING MODEL=$PWD/.lain/models/model.onnx For BGE-style asymmetric retrieval optimized for short natural-language queries , configure the prefix in .lain/tuning.toml : query prefix = "Represent this sentence for searching relevant passages: " Note: If no embedding model is configured, semantic search is filtered out of tools/list automatically so agents are never presented with an unusable tool. All graph, blast radius, navigation, and coordination tools function without an ONNX model. | Mode | Command | Use Case | |---|---|---| | stdio | lain mcp or lain server --transport stdio | Claude Code, Cursor, MCP clients | | http | lain server --transport http --port 9999 | Command Center dashboard + curl-driven JSON-RPC | - First-time setup issues? See QUICKSTART.md § First aid https://github.com/spuentesp/lain/blob/main/docs/QUICKSTART.md first-aid . - Run diagnostics : lain doctor reports repository and MCP readiness without changing files; lain doctor --json emits the same result for scripts and agents. - Hand-edit not picked up? The hot-reload watcher is non-recursive and uses atomic rename. Editing the file in place vim repos.yaml triggers a notify event within ~1 s. If you moved the file across directories, save it back into the project directory. - Repo stuck in indexing / degraded / unavailable ? The Command Center Overview tab displays these metrics live. curl -s -X POST http://localhost:9999/mcp \ -H 'Content-Type: application/json' \ -d '{"jsonrpc":"2.0","method":"tools/call","params":{"name":"get federation health","arguments":{}},"id":1}' - run build / run tests fail with "not found"? Editor-launched MCP servers inherit environments that lack version-manager shims rustup, nvm, pyenv, volta, asdf . LAIN searches known install locations automatically; to configure custom managers, see toolchains/README.md https://github.com/spuentesp/lain/blob/main/toolchains/README.md . - Answers look stale, or a symbol "doesn't exist" that clearly does? Check get health : - Build: shows the binary version and git SHA. MCP stdio processes are spawned once by the client and outlive source rebuilds; restart your editor/client to pick up a newly built binary. - Status: Degraded ⚠ indicates indexing timed out. Set LAIN REINDEX TIMEOUT=600 defaults to 300s outer, 60s per-repo to accommodate large working trees on cold cache. - Two agents not seeing each other? Verify they share the same workspace. State is persisted under ~/.local/lain/state/ , and list active agents confirms shared visibility. - Force re-indexing : Call request reload via MCP or touch repos.yaml . The hero demo recording in docs/screenshots/ can be re-recorded after UI changes: make record-demo Or run npm run record-demo --prefix tests/js for the Playwright driver and scripts/record-spa-demo.sh for ffmpeg encoding . For offline fixture testing, run make record-demo-small . MIT https://github.com/spuentesp/lain/blob/main/LICENSE — Copyright c 2026 spuentesp