Kivgraph
What it does
Cross-repository semantic code intelligence for AI coding agents, from a local code graph.
- Connects 在你的电脑上运行
- Account or key No key needed
- Category 开发者工具
安装
选择你使用的 AI 应用,然后复制设置。
通过 3/3 项检查 · 在你的电脑上运行
第一步:你用哪个 AI?
选择你用的应用,步骤会显示在下方。
可连接 Claude,但需要先准备
不需要密钥,不用安装程序 · 仅限电脑上的 Claude 应用(Windows 或 Mac),网页和手机不可用
双击文件,然后在电脑版 Claude 应用中点击 Install
试一试
用 Kivgraph 告诉我你能帮我做什么,然后做一件给我看✓ 如果 Claude 说正在使用 Kivgraph,或回答中出现 Kivgraph 的工具名称,就表示连接成功
菜单名称已于 2026年11月11日 对照 Anthropic 帮助页面核实 · 你的界面可能因版本略有不同
ChatGPT 暂时无法连接 Kivgraph
Kivgraph 需要安装在你的电脑上,而 ChatGPT 只能连接在网上运行的服务器
你可以改用这些方式
A · 使用能连接 Kivgraph 的应用
- 不需要密钥,不用安装程序 · 仅限电脑上的 Claude 应用(Windows 或 Mac),网页和手机不可用
B · 开发者工具 类中能连接 ChatGPT 的其他服务器
从同类中使用较多的里挑选,仅限用地址即可连接、无需密钥的
- Sentry✓ ChatGPT
- Compiler Explorer✓ ChatGPT
- GitDiagram✓ ChatGPT
Claude Code 暂时无法连接 Kivgraph
开发者只提供了电脑版 Claude 应用的安装文件
你可以改用这些方式
A · 使用能连接 Kivgraph 的应用
- 不需要密钥,不用安装程序 · 仅限电脑上的 Claude 应用(Windows 或 Mac),网页和手机不可用
Cursor 暂时无法连接 Kivgraph
开发者只提供了电脑版 Claude 应用的安装文件
你可以改用这些方式
A · 使用能连接 Kivgraph 的应用
- 不需要密钥,不用安装程序 · 仅限电脑上的 Claude 应用(Windows 或 Mac),网页和手机不可用
VS Code 暂时无法连接 Kivgraph
开发者只提供了电脑版 Claude 应用的安装文件
你可以改用这些方式
A · 使用能连接 Kivgraph 的应用
- 不需要密钥,不用安装程序 · 仅限电脑上的 Claude 应用(Windows 或 Mac),网页和手机不可用
一览
哪些应用可以使用它
依据这个服务器的连接方式和各应用自己的文档,检查于 2026年11月10日。
在你的电脑上运行,适用于
ClaudeGooseLM StudioMsty StudioCherry StudioVisual StudioWindsurf (Devin Desktop)ZedKiroClineContinueJetBrains AI AssistantCodex CLIGemini CLIOpenCodenodecodeWarpLibreChatRaycast
它能做什么
它在你的电脑上运行;安装后,工具列表会显示在你的 AI 应用中。
基础检查i
我们检查的是任何人都能看到的信息;我们没有阅读代码。不适用的检查项不计入总数。 · 检查于 2026年11月12日
- ✓ 通过
作者与仓库一致
注册名称已通过 GitHub 账户 Luqueee 验证,该账户拥有此仓库
- ✓ 通过
仍在维护
代码最后更新于今天
- ✓ 通过
有许可证
使用 Apache-2.0
- – 不适用
端点有响应
它在你的电脑上运行,没有互联网端点
- – 不适用
使用 OAuth 登录
在你的电脑上运行,改用本地设置
- … 尚未检查
工具描述中没有隐藏指令
它在你的电脑上运行;我们只能读取互联网端点的工具列表
讨论
提问、分享技巧或报告无法使用。按“有帮助”数量排序。
用户评价
还没有评价
与“开发者工具”中的其他服务器比较i
- 星标数多于此分类中其他 5816 个服务器的 93%
- 发布到注册中心的时间比此分类中其他 5816 个服务器的 33% 更新
- 基础检查结果等于或优于此分类中其他 5816 个服务器的 100%
与注册中心里这个分类中所有活跃的服务器比较。只有当发布者拥有该仓库时,星标数才计入。
常见问题
Kivgraph 安全吗?
它在 3 项适用的基础检查中通过了 3 项。我们没有阅读代码。在添加密钥或令牌之前,请先看看这个服务器要求哪些权限,并使用只具备所需权限的密钥。
它需要哪些密钥或代码?
注册中心没有列出必需的密钥,不过有些服务器会在首次使用时要求你登录。
它在我的电脑上运行还是在云端?
它在你的电脑上运行。你的 AI 应用会在需要时启动它,数据不会经过作者的服务器(除非它自己调用在线服务)。
它还在维护吗?
GitHub 仓库最后更新于今天,最新版本(0.9.10)于2个月前进入注册中心。
哪些应用可以使用它?
支持 MCP 的应用,例如 Claude Desktop、Claude Code、Cursor 和 VS Code。你的电脑上还需要有 MCPB。
有什么可以替代它?
同一分类中有:REA, GitHub, Serena MCP: the IDE for your agent
同类别的其他选择
同一分类中其他作者的服务器,按星标数排序。
如何看卡片:↓ 每周下载量i★ GitHub 星标i三大支柱:评价 · 检查 · 使用情况i更新 = 最新版本发布到注册中心的时间i
GitHub
Connect AI assistants to GitHub - manage repos, issues, PRs, and workflows through natural language.
Serena MCP: the IDE for your agent
A powerful toolkit for coding, providing semantic retrieval and editing capabilities.
作者的 README
打开 README
按原文从作者的仓库复制。PopNx MCP 没有审查或背书其内容。HTML、图片和脚本已被移除。链接会离开本站。
Kivgraph
[!NOTE]
Kivgraph began as a research project. It explored whether a local semantic
code graph could reduce the amount of context coding agents need to inspect.
The resulting tool remains useful for structured code navigation,
cross-repository relationships, and impact analysis. However, the research
**did not demonstrate a reliable net token saving over long, end-to-end agent
conversations**. While some isolated queries can use substantially fewer
tokens than
grepplus file reads, those savings did not consistently
translate into lower token usage across realistic long-running agent
workflows.
The benchmark results and their limitations are preserved in
benchmarks/, including the
Kivgraph is a local cross-repository code intelligence MCP server for AI coding agents. It builds a canonical semantic code graph across multiple registered repositories and answers questions about symbols, repository relationships, callers, dependencies and change impact.
Watch the Kivgraph UI demo video
kivgraph ui — read-only 3D view of the published graph.
It indexes a corpus once and serves an immutable graph: the edges are resolved by go/types, the TypeScript checker and rust-analyzer, not by matching names. That is the difference from a search tool, and it is what makes an empty answer worth something — an empty reference list means nobody calls it, not that nothing was found, and grep cannot tell those apart.
Kivgraph is focused on semantic code relationships, not automatic discovery of every HTTP, gRPC, Kafka or database runtime flow between services.
Documentation
Read the Kivgraph user documentation for installation, MCP clients, code intelligence, repository relationships and workspace code graphs. The same pages are the source of landing/src/content/docs in this checkout, which is what a reader on a fork or without a network still has.
What each tool answers
| the question | the tool |
|---|---|
| who calls this, what references this | find_references |
| who implements a type or method | find_implementations |
| what breaks if I change it | get_blast_radius |
| what does this reach outward | trace_dependencies |
| who uses it from another repository | find_cross_repo_consumers |
| where is it declared | find_symbol |
| I don't know what it is called, what files to open | find_by_intent |
| what is declared in this package | get_file_outline |
| give me the code of these symbols | get_source |
| everything about this one symbol | get_symbol |
| what is indexed, and is the graph current | list_repositories, graph_status |
| how is an asynchronous index progressing | get_index_status |
Thirteen read-only tools, plus two consent-gated mutations (index_project and start_index_project) that a client has to authorize before either can register a repository or publish a generation.
Every row that names a symbol carries its repository, path, qualified name and line range, so it can be opened without a second call, and every tool accepts that triple in place of an opaque key.
Where it loses. A rare name in one small repository is cheaper with grep, and indexing a small file costs more than reading it. It wins on common names, on transitive impact, on consumers in another repository, and on proving an absence. Measured over 29 questions against a 37-repository corpus (benchmarks/graph-tools-comparison/results-all.json, commit 954b9eb, tokenizer o200k_base): 35,961 tokens for Kivgraph against 267,980 for grep plus reading, both exact on 28 of the 29, median 5.95x per question in Kivgraph's favour. grep is cheaper on 5 of those 29, all of them at full recall on both sides: T1_go_trivial asks for a name the corpus declares twice, and there grep costs 0.53x what Kivgraph does.
A second harness, benchmarks/mcp-token-cost, compares against the host's own tool output captured verbatim, but it runs on Kivgraph's own single repository of 13,222 symbols: 7.64x on the answers themselves and 1.60x over a whole session, against a 2.41x floor set by the source bodies both arms pay for.
Status
Released and in use. kivgraph version reports the published release; the backlog and the acceptance gate of every phase are in TASKS.md.
- Languages: Go, TypeScript, Rust, Python and Dart. Python uses the bundled AST worker in fallback mode; those inferred references are
CANDIDATE, neverEXACT. Exact Python mode uses the bundled Pyright LSP adapter with an installed Pyright/BasedPyright server. Dart uses the Dart Analysis Server supplied by the Dart or Flutter SDK. - Semantic dependencies: Python and Dart imports can publish a package dependency when exactly one registered provider owns the requested package; symbol-level cross-repository edges require an explicit provider identity.
- Surface: thirteen read-only tools over STDIO, plus two consent-gated mutations (
index_projectandstart_index_project). The contract is docs/protocol/mcp-surface-v3.md. - Storage: LadybugDB is canonical; queries are served from an immutable HotSnapshot published atomically, never from the database.
- Platforms:
linux/amd64,darwin/arm64andwindows/amd64. - Viewer:
kivgraph uiserves a read-only 3D view of the published graph.
Installation
Install the MCP with one script
The installer detects the platform, downloads the latest published MCP release for it, verifies both the release archive and the bundle checksums, and installs it without requiring Go or pnpm. The release contains the Go server, the pinned LadybugDB library, the TypeScript worker, the bundled Python AST worker, the pinned rust-analyzer, the grammar manifest and the web viewer, whose assets are 2.3 MB of the bundle. scripts/build-bundle.sh --mcp-only produces a bundle without the viewer for anyone who wants one. --slim goes further for anyone packaging an .mcpb: it leaves out the pinned rust-analyzer and every symbol a debugger would read, which is 46.3 MB packaged against 24.9 MB. It downloads nothing later, so that bundle reads Rust only where the machine already has an analyzer on its PATH.
Published bundles: Linux amd64, macOS arm64 and Windows amd64.
Runtime requirements: Bash on Linux and macOS or PowerShell 5.1 or later on Windows, Node.js 22 or later, Python 3.10 or later when indexing Python, and on the POSIX platforms curl, tar, and sha256sum or shasum. The bundle carries its own rust-analyzer; indexing Rust repositories additionally needs cargo on the PATH, and indexing Dart needs the Dart or Flutter SDK. On Windows the installer also installs the Visual C++ redistributable, without which kivgraph.exe does not start.
On macOS the binaries are not notarized. A release downloaded with curl is not quarantined and runs; a copy downloaded with a browser needs xattr -dr com.apple.quarantine. See docs/development/macos.md.
Install the latest release in one command. On Linux and macOS the same line covers both, because the installer reads uname and picks its own archive:
curl -fsSL https://kivgraph.dev/install.sh | bash
On Windows, where install.sh cannot run because there is no POSIX shell:
irm https://kivgraph.dev/install.ps1 | iex
install.ps1 is a second implementation of the same pre-extraction checks, and internal/release/install_parity_test.go fails when either script grows a check the other lacks. Piping it into Invoke-Expression turns its #Requires -Version 5.1 into a comment; download it to a file and run it as one to keep that guard.
From a checkout, either installer can be run directly:
./scripts/install.sh
To install a specific release instead of the latest one:
KIVGRAPH_VERSION=v0.9.10 ./scripts/install.sh
The script installs the bundle in ~/.local/opt/kivgraph and puts launchers in ~/.local/bin; on Windows it is %LOCALAPPDATA%\Programs\kivgraph and %LOCALAPPDATA%\Programs\kivgraph-bin. It never modifies a registered repository, creates an index, or replaces configuration files. To use a different location, set KIVGRAPH_INSTALL_ROOT and KIVGRAPH_BIN_DIR.
Add the launcher directory to the current shell and verify both runtimes:
export PATH="$HOME/.local/bin:$PATH"
kivgraph version kivgraph-ts-worker <<'EOF' hello EOF
Check for a newer release or update the installed bundle:
kivgraph update --check
kivgraph update
Bundle replacement is atomic, preserves the configuration and graph state, verifies the release and bundle checksums, and replaces the installed bundle. The post-install runtime refresh may partially complete, fail, and make the command exit non-zero. It also restarts an installed supervised daemon and refreshes Kivgraph-managed user hooks, skills and MCP registrations. A stale supervisor returns an error and is not restarted. Missing, foreign and project-scoped integrations are left alone. Client-owned serve and ui processes still need a restart, or --stop, to use the new binary.
Development builds use a separate prerelease channel. Install one explicitly, then select that channel for later checks:
release=vX.Y.Z-dev.N
curl -fsSL \ "https://github.com/Luqueee/kivgraph/releases/download/$release/install.sh" | KIVGRAPH_VERSION="$release" bash kivgraph update --channel dev
For a prerelease binary, omitting --channel already follows dev; stable installations continue to follow the stable channel. KIVGRAPH_UPDATE_CHANNEL can be used instead of the flag, including for the interactive update notice.
To remove the installed bundle and launchers without deleting configuration, repository registrations or graph state:
uninstall_url=https://github.com/Luqueee/kivgraph/releases/latest/download/uninstall.sh
curl -fsSL "$uninstall_url" -o /tmp/kivgraph-uninstall.sh bash /tmp/kivgraph-uninstall.sh
Use bash /tmp/kivgraph-uninstall.sh --yes for a non-interactive removal. Windows users can run the corresponding uninstall.ps1 with PowerShell.
When kivgraph is invoked without a command from an interactive terminal, it checks for a newer release with an 800 ms timeout and a 24-hour cache in the platform cache directory ($XDG_CACHE_HOME on Linux and $HOME/Library/Caches on macOS), under kivgraph/update-check.json. The optional check never blocks the command when the network is unavailable.
Interactive command output uses semantic ANSI colors when the destination is a terminal. Set NO_COLOR or redirect output to keep it plain.
Configure Kivgraph and its coding agents
After an interactive release installation, the installer asks whether it should run this same guided setup. Run it later, or run it directly from a checkout:
kivgraph configure
configure opens one selector for the coding agents detected on this machine and installs each user-scoped surface that the selected agent supports. It also adds user-level Kivgraph instructions for every selected agent. It initializes the empty Kivgraph configuration when needed, but it does not register a repository or run an index. Repeat --target TARGET for scripted setup; omit it to open the selector.
Use ↑/↓ (or j/k) to move, space to toggle an agent, a to select all, n to select none, Enter to confirm, and q or Esc to cancel. The daemon is offe