# codemesh [Health: Active] **Category:** πŸ’» Developer Tools **Repository:** https://github.com/pyalwin/codemesh **GitHub Stars:** 0 **npm Downloads (last month):** 74 **Views:** 0 **Installs:** 0 **Upvotes:** 0 **Directory Page:** https://allmcps.com/mcp/codemesh ## Description Intelligent code knowledge graph for AI coding agents β€” 71% cheaper, 72% faster ## Claude Desktop Quick Installation Install path detected from listing signals. Uses `npx` (confidence: high): ```json "mcpServers": { "codemesh": { "command": "npx", "args": ["-y","@pyalwin/codemesh"] } } ``` ## Documentation & README
# Codemesh **Intelligent code knowledge graph for AI coding agents** **71% cheaper, 72% faster, 82% fewer tool calls** vs baseline Grep+Read\ on 6 real-world repos (Sonnet 4.6) β€” from a single `codemesh index`. [![npm](https://img.shields.io/npm/v/@pyalwin/codemesh)](https://www.npmjs.com/package/@pyalwin/codemesh) [![Tests](https://img.shields.io/badge/tests-124%20passed-brightgreen)]() [![License](https://img.shields.io/badge/license-MIT-blue)]() [![TypeScript](https://img.shields.io/badge/TypeScript-5.7-blue)]() [![MCP](https://img.shields.io/badge/MCP-compatible-purple)]() [Benchmarks](#benchmarks) · [Quick Start](#quick-start) · [Integrations](#client-integrations) · [Write-Back](#agent-write-back-the-graph-that-gets-smarter) · [How It Works](#how-it-works) · [API Reference](#mcp-tools) · [Full Results](https://github.com/pyalwin/codemesh/blob/HEAD/docs/benchmark-results.md)
--- ## The Problem AI coding agents waste **40-80% of their tokens on discovery** β€” grepping through files, reading irrelevant code, and rebuilding context they've already seen in previous sessions. On a 600-file codebase, a typical exploration task involves 10+ file reads before the agent even knows what's relevant. ``` Before: Agent β†’ Grep β†’ 50 matches β†’ Read 10 files β†’ Understand β†’ Work After: Agent β†’ codemesh_explore β†’ 3 relevant files β†’ codemesh_trace β†’ full path β†’ Work ``` Codemesh is an MCP server that gives agents a persistent, queryable knowledge graph. The graph gets smarter over time: agents write back what they learn, so the next session starts informed. --- ## Benchmarks Benchmarked on 6 real-world codebases (Alamofire, Excalidraw, VS Code, Swift Compiler, pydantic-validators, pydantic-basemodel) with Claude Sonnet 4.6, compared alongside baseline and graph-based approaches for context. Full methodology, per-repo breakdowns, and pairwise comparisons: [`docs/benchmark-results.md`](https://github.com/pyalwin/codemesh/blob/HEAD/docs/benchmark-results.md) | [Early pydantic evals](https://github.com/pyalwin/codemesh/blob/HEAD/docs/experiments/pydantic-eval-results.md) ### Cost | Mode | Alamofire | Excalidraw | VS Code | Swift Compiler[^swift] | pydantic-validators | pydantic-basemodel | **Avg** | |---|---:|---:|---:|---:|---:|---:|---:| | Baseline | $0.54 | $0.89 | $0.21 | $0.83 | $1.32 | $0.78 | $0.76 | | **Codemesh MCP** | **$0.25** | **$0.21** | **$0.16** | **$0.23** | **$0.33** | **$0.13** | **$0.22** | | Codemesh CLI | $0.67 | $0.51 | $0.16 | $0.83 | $1.00 | $0.18 | $0.56 | | Codegraph | $0.37 | $0.56 | $0.57 | $0.74 | $0.29 | $0.19 | $0.45 | ### Time | Mode | Alamofire | Excalidraw | VS Code | Swift[^swift] | pydantic-v | pydantic-b | **Avg** | |---|---:|---:|---:|---:|---:|---:|---:| | Baseline | 180s | 191s | 87s | 199s | 352s | 232s | 207s | | **Codemesh MCP** | **78s** | **45s** | **35s** | **87s** | **72s** | **32s** | **58s** | | Codemesh CLI | 226s | 177s | 62s | 227s | 235s | 51s | 163s | | Codegraph | 134s | 180s | 192s | 199s | 75s | 60s | 140s | ### Tool calls (agent turns) | Mode | Alamofire | Excalidraw | VS Code | Swift[^swift] | pydantic-v | pydantic-b | **Avg** | |---|---:|---:|---:|---:|---:|---:|---:| | Baseline | 31 | 48 | 12 | 29 | 84 | 65 | 45 | | **Codemesh MCP** | **9** | **5** | **3** | **14** | **14** | **3** | **8** | | Codemesh CLI | 30 | 32 | 12 | 56 | 64 | 9 | 34 | | Codegraph | 31 | 35 | 44 | 44 | 20 | 12 | 31 | ### Quality (1–10, LLM-as-judge) | Mode | Alamofire[^alamo] | Excalidraw | VS Code | Swift Compiler | pydantic-validators | pydantic-basemodel | **Avg** | |---|---:|---:|---:|---:|---:|---:|---:| | Baseline | n/a | 9 | 8 | 7 | 2 | 9 | 7.0 | | **Codemesh MCP** | 9 | 9 | 7 | 8 | 7 | 7.8 | **7.9** | | Codemesh CLI | 9 | 7 | 7 | 9 | 1 | 8.4 | 6.9 | | Codegraph | 8 | 9 | 8.7 | 8 | 8 | 9 | **8.4** | ### Cost savings: Codemesh MCP vs Baseline | Repo | Baseline | Codemesh MCP | **Cost saved** | **Time saved** | |---|---:|---:|---:|---:| | Alamofire | $0.54 | $0.25 | **βˆ’54%** | **βˆ’57%** (180s β†’ 78s) | | Excalidraw | $0.89 | $0.21 | **βˆ’76%** | **βˆ’76%** (191s β†’ 45s) | | VS Code | $0.21 | $0.16 | **βˆ’24%** | **βˆ’60%** (87s β†’ 35s) | | Swift Compiler[^swift] | $0.83 | $0.23 | **βˆ’72%** | **βˆ’56%** (199s β†’ 87s) | | pydantic-validators | $1.32 | $0.33 | **βˆ’75%** | **βˆ’79%** (352s β†’ 72s) | | pydantic-basemodel | $0.78 | $0.13 | **βˆ’83%** | **βˆ’86%** (232s β†’ 32s) | | **Average** | **$0.76** | **$0.22** | **βˆ’71%** | **βˆ’72%** | > [!NOTE] > **Codemesh MCP** achieves the lowest cost and fastest time of any mode tested β€” **71% cheaper and 72% faster than baseline** on average across 6 repos, using **82% fewer tool calls** (8 vs 45). Quality is comparable to baseline (7.9 vs 7.0); Codegraph edges Codemesh on quality (8.4) but at roughly double the cost ($0.45 vs $0.22). Every repo shows cost and time savings β€” including the comprehension-heavy queries (Excalidraw, pydantic-basemodel) that regressed in prior builds of codemesh. [^swift]: https://github.com/pyalwin/codemesh/blob/HEAD/Swift Compiler's codemesh index failed to complete (indexer regression on 30k+ file codebases β€” see [known issues](https://github.com/pyalwin/codemesh/blob/HEAD/docs/benchmark-results.md)). The codemesh numbers above reflect agent behavior with an empty retrieval graph, falling back to Read + LSP β€” still ahead of baseline, but unrepresentative of codemesh's capability on a properly-indexed Swift repo. [^alamo]: https://github.com/pyalwin/codemesh/blob/HEAD/Baseline for Alamofire hit a judge error (score recorded as 0 but not meaningful); excluded from the Baseline average. --- ## Quick Start ### 1. Install ```bash npm install -g @pyalwin/codemesh ``` Or run directly without installing: ```bash npx -y @pyalwin/codemesh --help ```
Build from source ```bash git clone https://github.com/pyalwin/codemesh.git cd codemesh npm install && npm run build npm link ```
> Verify the install: `codemesh --version` should print the package version. ### 2. Index your project ```bash cd /your/project codemesh index --with-embeddings ``` ``` Indexed 656 files Symbols found: 16733 Edges created: 33266 Duration: 10009ms PageRank: 13843 nodes scored Embeddings: 13187 symbols embedded ``` ### 3. Choose your mode Codemesh offers two ways to integrate with AI agents:
Option A: MCP Server (structured tool calls) Add to your Claude Code MCP config (`~/.claude/mcp-servers.json` or project `.mcp.json`): ```json { "mcpServers": { "codemesh": { "command": "npx", "args": ["-y", "@pyalwin/codemesh"], "env": { "CODEMESH_PROJECT_ROOT": "/path/to/your/project" } } } } ``` The agent gets native MCP tools: - `codemesh_answer` β€” one-call question answering (PRIMARY) - `codemesh_explore` β€” search, context (multi-target), impact - `codemesh_trace` β€” follow call chains - `codemesh_enrich` / `codemesh_workflow` β€” write back - `codemesh_status` β€” health check **Best for:** Opus, structured workflows, enrichment/write-back
Option B: CLI Mode (via Bash β€” zero MCP overhead) No MCP config needed. The agent calls codemesh directly via Bash: ```bash export CODEMESH_PROJECT_ROOT=/path/to/your/project # Primary β€” one-call question answering: codemesh explore answer "How does request handling work?" # Follow-up commands: codemesh explore search "request flow" codemesh explore context Source/Core/Session.swift Source/Core/Request.swift codemesh explore trace Session.request --depth 5 codemesh explore semantic "network request handling" # requires --with-embeddings ``` All commands return JSON to stdout. No MCP server process, no protocol overhead. **Best for:** Sonnet/Haiku, speed-sensitive workflows, simpler setup
### Which mode should I use? | | MCP Server | CLI Mode | |---|---|---| | **Setup** | MCP config file | Just `export CODEMESH_PROJECT_ROOT` | | **Overhead** | MCP protocol per call | Zero β€” direct subprocess | | **Enrichment** | Native `codemesh_enrich` tool | Via `Bash("codemesh enrich ...")` | | **Best model** | Opus (follows MCP well) | **Sonnet** (55% cheaper, 61% faster than baseline) | | **Recommended** | Complex codebases | **Default choice** | ### 4. Use it The agent now has 6 new tools. Query the graph before reading code: ``` You: "Find how pydantic handles validation" Agent calls: codemesh_answer({ question: "How does pydantic handle validation?" }) gets: 9 relevant files ranked by PageRank, call chains, git hotspots, co-change relationships, 5 suggested reads Agent calls: Read("pydantic/functional_validators.py", lines 1-50) reads: only the specific lines suggested by the answer tool Agent calls: codemesh_enrich({ path: "pydantic/functional_validators.py", summary: "Primary V2 validator API..." }) saves: summary for next session ``` --- ## Client Integrations Codemesh speaks the Model Context Protocol, so any MCP-compatible client can use it. Paste one of the snippets below, restart the client, and the six `codemesh_*` tools show up in the agent's toolbox.
Claude Code (CLI) Add to `~/.claude/mcp-servers.json` (user-wide) or `.mcp.json` (project-local): ```json { "mcpServers": { "codemesh": { "command": "npx", "args": ["-y", "@pyalwin/codemesh"], "env": { "CODEMESH_PROJECT_ROOT": "/absolute/path/to/your/project" } } } } ```
Claude Desktop (macOS / Windows app) Edit `claude_desktop_config.json`: - **macOS:** `~/Library/Application Support/Claude/claude_desktop_config.json` - **Windows:** `%APPDATA%\Claude\claude_desktop_config.json` ```json { "mcpServers": { "codemesh": { "command": "npx", "args": ["-y", "@pyalwin/codemesh"], "env": { "CODEMESH_PROJECT_ROOT": "/absolute/path/to/your/project" } } } } ``` Restart Claude Desktop. Codemesh's tools will appear in the tool picker (hammer icon).
Cursor β€” stop the agent from wandering your codebase Cursor reads `.cursor/mcp.json` per project (or `~/.cursor/mcp.json` for all projects): ```json { "mcpServers": { "codemesh": { "command": "npx", "args": ["-y", "@pyalwin/codemesh"], "env": { "CODEMESH_PROJECT_ROOT": "${workspaceFolder}" } } } } ``` Open **Settings β†’ MCP**, confirm `codemesh` is green, then mention it in a prompt (`@codemesh how does auth work?`) to nudge the agent toward graph queries instead of recursive Grep.
Windsurf / VS Code (Continue) Add to `~/.continue/config.json` under `experimental.modelContextProtocolServers`: ```json { "experimental": { "modelContextProtocolServers": [ { "transport": { "type": "stdio", "command": "npx", "args": ["-y", "@pyalwin/codemesh"], "env": { "CODEMESH_PROJECT_ROOT": "/absolute/path/to/your/project" } } } ] } } ```
--- ## Agent Write-Back: the graph that gets smarter Every other code-intelligence tool indexes your repo once and hands the agent a read-only view. Codemesh lets the agent **teach the graph** as it works β€” summaries, workflows, and cross-concept links persist across sessions and survive re-indexing. ```ts // Session 1 β€” agent reads unfamiliar code, then writes back what it learned. codemesh_enrich({ path: "pydantic/functional_validators.py", summary: "Primary V2 validator API. `@field_validator` wraps " + "`_decorators.FieldValidatorDecoratorInfo`; `mode='before'|'after'` " + "toggles pre/post-coercion execution. Extends BaseValidator.", concepts: ["validation", "decorators", "v2-api"] }) // Session 1 β€” agent traces a multi-file flow, records the path. codemesh_workflow({ name: "pydantic field validation", description: "Request β†’ BaseModel.__init__ β†’ SchemaValidator β†’ field_validator", files: [ "pydantic/main.py", "pydantic/_internal/_model_construction.py", "pydantic/functional_validators.py" ] }) // Session 2 (days later) β€” same question, different agent instance. codemesh_answer({ question: "How does pydantic validate fields?" }) // β†’ returns the enriched summary AND the 3-file workflow from Session 1 // before the agent reads a single line. Zero rediscovery cost. ``` The graph now knows things no static analyzer could infer: why a file matters, which files move together, what a maintainer called a concept. Re-indexing rebuilds the structural layer (files, symbols, imports, calls) but **preserves every enrichment** β€” entries only go stale when their referenced files change. See `codemesh_enrich` and `codemesh_workflow` under [MCP Tools](#mcp-tools). --- ## How It Works ``` β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β” β”‚ Knowledge Graph β”‚ β”‚ β”‚ β”‚ β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β” β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β” β”‚ β”‚ β”‚Structuralβ”‚ β”‚ Semantic β”‚ β”‚ β”‚ β”‚ (auto) β”‚ β”‚ (agents) β”‚ β”‚ β”‚ β”‚ β”‚ β”‚ β”‚ β”‚ β”‚ β”‚ files β”‚ β”‚ summaries β”‚ β”‚ β”‚ β”‚ symbols β”‚ β”‚ workflows β”‚ β”‚ β”‚ β”‚ imports β”‚ β”‚ concepts β”‚ β”‚ β”‚ β”‚ calls β”‚ β”‚ enrichments β”‚ β”‚ β”‚ β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜ β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜ β”‚ β”‚ β”‚ β”‚ β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β” β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β” β”‚ β”‚ β”‚ Git β”‚ β”‚ Search β”‚ β”‚ β”‚ β”‚ Intel β”‚ β”‚ β”‚ β”‚ β”‚ β”‚ β”‚ β”‚ FTS5 (exact) β”‚ β”‚ β”‚ β”‚ hotspots β”‚ β”‚ Trigram (fuzzy)β”‚ β”‚ β”‚ β”‚ co-changeβ”‚ β”‚ LanceDB (sem) β”‚ β”‚ β”‚ β”‚ churn β”‚ β”‚ PageRank β”‚ β”‚ β”‚ β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜ β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜ β”‚ β”‚ β”‚ β”‚ SQLite + LanceDB β”‚ β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”¬β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜ β”‚ β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”΄β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β” β”‚ MCP Server / CLI (7 tools) β”‚ β”‚ β”‚ β”‚ answer Β· explore Β· trace β”‚ β”‚ enrich Β· workflow Β· status β”‚ β””β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”˜ ``` **Structural layer** (automatic) β€” Tree-sitter parses your code into files, symbols (functions, classes, methods), and relationships (imports, calls, extends). Rebuilt on each index. **Semantic layer** (agent-built) β€” As agents work with your code, they write back summaries and workflow paths. These survive re-indexing and accumulate across sessions. Invalidated when referenced files change. --- ## MCP Tools | Tool | Purpose | Example | |---|---|---| | `codemesh_answer` | **One-call context assembly** β€” returns all relevant files, call chains, hotspots, suggested reads | `codemesh_answer({ question: "How does auth work?" })` | | `codemesh_explore` | Search, context (multi-target), impact analysis | `codemesh_explore({ action: "search", query: "auth" })` | | `codemesh_trace` | Follow call chains with source code | `codemesh_trace({ symbol: "login", depth: 5 })` | | `codemesh_enrich` | Write back what you learned for future sessions | `codemesh_enrich({ path: "src/auth.py", summary: "..." })` | | `codemesh_workflow` | Record multi-file workflow paths | `codemesh_workflow({ name: "login flow", files: [...] })` | | `codemesh_status` | Graph health check | `codemesh_status()` | --- ## CLI ```bash codemesh index # structural + git intel + pagerank codemesh index --with-embeddings # + semantic vectors (~80MB model, zero API cost) codemesh status # graph statistics codemesh rebuild # purge and re-index codemesh explore answer "question" # one-call context assembly (PRIMARY) codemesh explore search "query" # FTS5 + trigram + semantic search codemesh explore context file1 file2 # multi-target context codemesh explore trace symbol --depth 5 # follow call chains codemesh explore semantic "query" # vector similarity (needs embeddings) codemesh explore impact file # reverse dependencies ``` --- ## Optional: Hooks & Skills
Skill β€” teaches agents the graph-first workflow Copy `skills/codemesh.md` to `~/.claude/skills/` or your project's `.claude/skills/`. ```bash # Install the skill so Claude Code loads the workflow automatically cp /path/to/codemesh/skills/codemesh.md /your/project/.claude/skills/ ``` The skill instructs agents to query the graph before using Grep/Read, and to write back via `codemesh_enrich` after reading code.
Hooks β€” automatic pre-read context injection Add to `.claude/settings.json`: ```json { "hooks": { "pre_tool_use": [{ "matcher": "Read", "command": "/path/to/codemesh/hooks/pre-read.sh" }], "post_tool_use": [{ "matcher": "Read", "command": "/path/to/codemesh/hooks/post-read.sh" }] } } ``` - **Pre-read** β€” Injects cached summaries before file reads - **Post-read** β€” Nudges the agent to enrich after reading unfamiliar files
--- ## Supported Languages
TypeScriptJavaScriptPythonGoRustJavaC#
RubyPHPCC++SwiftKotlinDart
Any language with a [tree-sitter](https://tree-sitter.github.io/tree-sitter/) grammar can be added. --- ## Graph Data Model ### Nodes | Type | Source | Key Fields | |:--|:--|:--| | `file` | Static (tree-sitter) | `path`, `hash`, `last_indexed_at` | | `symbol` | Static (tree-sitter) | `name`, `kind`, `file_path`, `line_start`, `line_end`, `signature` | | `concept` | Agent-written | `summary`, `last_updated_by`, `stale` | | `workflow` | Agent-written | `description`, `file_sequence`, `last_walked_at` | ### Edges | Type | Direction | Source | |:--|:--|:--| | `contains` | file → symbol | Static | | `imports` | file → file | Static | | `calls` | symbol → symbol | Static | | `extends` | symbol → symbol | Static | | `describes` | concept → file/symbol | Agent | | `related_to` | concept → concept | Agent | | `traverses` | workflow → file | Agent | --- ## Architecture ``` codemesh/ β”œβ”€β”€ src/ β”‚ β”œβ”€β”€ index.ts # MCP server entry (stdio transport) β”‚ β”œβ”€β”€ server.ts # Tool registration (zod schemas) β”‚ β”œβ”€β”€ graph/ β”‚ β”‚ β”œβ”€β”€ types.ts # Node/edge type definitions β”‚ β”‚ β”œβ”€β”€ storage.ts # StorageBackend interface (swappable) β”‚ β”‚ └── sqlite.ts # SQLite + FTS5 implementation β”‚ β”œβ”€β”€ indexer/ β”‚ β”‚ β”œβ”€β”€ indexer.ts # File walking, hashing, incremental indexing β”‚ β”‚ β”œβ”€β”€ parser.ts # Tree-sitter AST extraction β”‚ β”‚ └── languages.ts # Language registry (ext β†’ grammar) β”‚ β”œβ”€β”€ tools/ # 6 MCP tool handlers β”‚ └── cli.ts # CLI entry point β”œβ”€β”€ skills/codemesh.md # Agent education skill β”œβ”€β”€ hooks/ # Pre/post read hooks └── eval/ # Eval framework (5 tasks, 3 models) ``` Storage is **backend-agnostic**. The `StorageBackend` interface abstracts all persistence. v1 uses SQLite with FTS5 for zero-dependency local operation. The interface supports swapping to Memgraph, Neo4j, or other graph databases. --- ## Eval Framework Reproducible evaluation harness with LLM-as-judge scoring: ```bash # Setup npm install -g @pyalwin/codemesh git clone --depth 1 https://github.com/Alamofire/Alamofire.git /tmp/alamofire # ... clone other repos ... # Index CODEMESH_PROJECT_ROOT=/tmp/alamofire codemesh index # Run benchmarks python3 eval/head_to_head.py --model sonnet alamofire excalidraw vscode swift-compiler ``` See [`docs/benchmark-results.md`](https://github.com/pyalwin/codemesh/blob/HEAD/docs/benchmark-results.md) for full methodology and results. Early pydantic evals are archived in [`docs/experiments/`](https://github.com/pyalwin/codemesh/blob/HEAD/docs/experiments/). --- ## vs. Existing Tools | Feature | [CodeGraph](https://github.com/colbymchenry/codegraph) | [Graphify](https://github.com/safishamsi/graphify) | [Axon](https://github.com/harshkedia177/axon) | **Codemesh** | |:--|:--:|:--:|:--:|:--:| | Structural indexing | Yes | Yes | Yes | Yes | | FTS search | Yes | — | Yes | Yes | | Agent write-back | — | — | — | **Yes** | | Workflow memory | — | — | — | **Yes** | | Hook interception | — | — | — | **Yes** | | Backend-swappable | — | — | — | **Yes** | | Eval framework | — | — | — | **Yes** | | Published benchmarks | — | — | — | **Yes** | --- ## Development ```bash bun install # Install dependencies bun run build # Compile TypeScript bun run test # Run 102 tests bun run dev # Watch mode bun run lint # Type check ``` --- ## Contributing Contributions welcome. Areas for improvement: - **More languages** — Add tree-sitter grammars and language-specific extractors - **AST-diff invalidation** — Function-level instead of file-level staleness detection - **Graph backends** — Memgraph/Neo4j adapters for `StorageBackend` - **Semantic search** — Embedding columns alongside FTS5 - **Agent adoption** — Better patterns for agents to prefer graph tools naturally --- ## License MIT