LayerMap

io.github.coffeecoprojectv0.1.4更新於 Oct 5, 2026

Call graph for coding agents: callers up to HTTP routes, callees and usages for TS, Go, Python, Java

已驗證STDIO僅桌面Developer ToolsKnowledge & Memory

概覽

AI 產生的概覽

LayerMap 為編碼代理提供程式碼庫的唯讀呼叫圖,可沿呼叫鏈向上追溯到 HTTP 路由,並分析 TypeScript、JavaScript、Go、Python 與 Java 的呼叫方與被呼叫方。

功能
LayerMap 會在本機為程式碼庫建立對應,並提供三個唯讀工具:project_explore_map 用來查看目錄、檔案或宣告的呼叫方與被呼叫方(direction 為 INCOMING 或 OUTGOING,depth 最多 8 層),project_search_map 依名稱、路徑或文件中的詞彙尋找宣告,project_find_references 找出某個宣告的所有使用位置。呼叫關係由各語言本身的編譯器或型別檢查器解析,因此介面、基底類別、樣板與裝飾器的呼叫都能依編譯器的解析結果呈現。對應會在每次呼叫時增量更新,並快取在儲存庫之外。
適用情境
當代理需要在程式碼庫中回答影響範圍問題(例如某處變更會影響哪些 HTTP 路由或處理函式),而不想進行大量分散搜尋時,適合使用。適用於重視編譯器層級呼叫方與被呼叫方準確性的 TypeScript、JavaScript、Go、Python 與 Java 專案。
執行需求
透過 npx(npm 套件 layermap)以 stdio 在本機執行。需要 Node.js 22.22 或更新版本、macOS 或 Linux(x64 或 arm64),以及一個 Git 儲存庫;Java 專案還需要 JDK 21 或更新版本。未宣告帳號、API 金鑰或環境變數;可選環境變數 LAYERMAP_CACHE 用來指定快取位置。
安裝前請注意
這些工具是唯讀的,README 說明 LayerMap 完全在本機執行且不向任何地方傳送資料。安裝指令可能會註冊此伺服器、允許其唯讀工具無需確認即可執行,並在代理的指令檔中加入一句話;團隊使用可用 --scope project,撤銷可用 remove 指令。大型儲存庫首次建立對應可能需要數分鐘。

安裝

在 SourceWeft 中

  1. 開啟 儀表板中的 LayerMap,將其新增到工作區。
  2. 為需要使用其工具的對話啟用該服務。

Desktop only,透過 STDIO。 STDIO 服務會啟動本機處理程序,因此需要 SourceWeft 桌面主機。

其他 MCP 客戶端

參照 儲存庫 中的啟動說明。

README

[LayerMap]

LayerMap

English · 中文

[npm] [License]

[Claude Code asks which HTTP endpoints changing Storage.MarkFeedAsRead affects; one LayerMap call returns all four callers with their routes]

A real Claude Code session on Miniflux, replayed: one map call finds every endpoint, then the agent confirms them in source.

A layered map of your codebase for coding agents. LayerMap gives Claude Code, Codex, DeepSeek Harness and other MCP clients three read-only tools that answer, in one call, what an agent otherwise works out with dozens of searches:

  • What calls this, and what does changing it affect? Callers traced up to 8 levels, to the HTTP routes, handlers, jobs and commands they start from.
  • What does it call? Callees down to 8 levels.
  • What is here? Modules, each file's declarations, and every usage of a symbol.

It maps TypeScript, JavaScript, Go, Python and Java with each language's own compiler, so calls through interfaces, base classes, templates and decorators resolve as the compiler resolves them, including the ones text search misses.

Example

In Miniflux, one call on Storage.MarkFeedAsRead finds all four ways a request reaches it, each with its route:

DECLARATION internal/storage/entry.go: Storage.MarkFeedAsRead m656-680 exported  ← called by internal/api/feed.go: handler.markFeedAsRead m140-155 @149  ← called by internal/fever/handler.go: handler.handleWriteFeeds m492-519 @508  ← called by internal/googlereader/handler.go: handler.markAllAsReadHandler m1261-1338 @1311  ← called by internal/ui/feed_mark_as_read.go: handler.markFeedAsRead m14-30 @24[1] internal/api/feed.go: handler.markFeedAsRead m140-155  ← used as value by internal/api/api.go: Serve f25-84 @61 "/feeds/{feedID}/mark-all-as-read"[1] internal/googlereader/handler.go: handler.markAllAsReadHandler m1261-1338  ← used as value by internal/googlereader/handler.go: Serve f44-64 @62 "/mark-all-as-read"[1] internal/ui/feed_mark_as_read.go: handler.markFeedAsRead m14-30  ← used as value by internal/ui/ui.go: Serve f18-180 @77 "/feed/{feedID}/mark-all-as-read"… (the Fever API, then on down to main.go: main)

m and f mark methods and functions with their lines; @ is the line of the call, and the quoted path is the route it is registered under.

Results

Impact questions ("which HTTP endpoints does changing this affect?") on public Go, Python and Java projects, graded blind against truth sets cross-checked with each language's own toolchain:

With LayerMapWithout
Endpoints found, at most 15 requests (gpt-5.5, 6 tasks × 3 runs)97.9%58.8%
Used unprompted by Claude Code / Codex (plugin installed)6 of 6 / 6 of 6—
Endpoints found, no request limit (Claude Code / Codex)99.3% / 98.0%99.0% / 98.6%
Cost, no request limit (Claude Code in USD / Codex in input tokens)−32% / −55%
Time, no request limit+24% / +17%

With a tight budget, the map finds far more of the affected code, at about 24% more tokens. With no limit, both agents get there either way. The map makes the run cheaper, and the agents explore more widely, which takes longer. These are small samples on tasks written by LayerMap's authors. The report has the setup, the published tasks and the limitations.

How LayerMap differs

Other tools give agents part of this. LayerMap combines compiler accuracy with whole-chain traces in one local map:

Tree-sitter code graphs (codegraph, GitNexus, …)Language servers and IDEs (Claude Code's LSP tool, Serena, JetBrains)Embedding search (Claude Context, Augment)LayerMap
How calls are linkedMatched by names, imports and framework rules, often with confidence scoresThe compiler's own resolutionNot linked; similar code is retrievedResolved by each language's compiler or type checker
Callers traced to routes and handlersIn several tools, at varying depthOne level per request (JetBrains: 5 by default)—Up to 8 levels in one call
Kept as a mapYes, with file watchingNo, answered live by a running serverAn index of code chunksYes, brought up to date on every call
Runs on your machineYes; some send anonymous telemetryYesUsually with cloud embeddingsYes, and sends nothing

Another tool fits better if you need:

  • more languages, since tree-sitter graphs cover 30 or more;
  • Windows;
  • renaming and refactoring, from language servers;
  • search by meaning, from embedding search;
  • search across many repositories, with Sourcegraph.

Install

Claude Code

/plugin marketplace add coffeecoproject/layermap/plugin install layermap@layermap

Codex

codex plugin marketplace add coffeecoproject/layermapcodex plugin add layermap@layermap

DeepSeek Harness

npx layermap setup dsh

This adds LayerMap to every dsh profile. Each session maps the project dsh was started in, and npx layermap remove dsh undoes it.

In every agent, start a new session in a Git repository and ask as usual. LayerMap tells the agent when the map helps. On first launch, npx downloads the pinned layermap package from npm.

Claude Code asks once before each map tool runs in a project. Choose "don't ask again", or run npx layermap allow claude to allow the plugin's read-only tools everywhere.

Without plugins: run npx layermap setup claude or npx layermap setup codex. Setup does three things:

  • registers the server;
  • lets its read-only tools run without a prompt;
  • adds one marked sentence to the agent's instructions file.

Use --scope project to set it up for a whole team, and npx layermap remove … to undo it. For any other MCP client, run npx -y layermap mcp as a stdio server in the repository.

Requirements:

  • Node.js 22.22 or later.
  • macOS or Linux (x64 or arm64).
  • A Git repository.
  • For Java projects, a JDK 21 or later.

How it works

The first call in a repository builds its map: seconds for a small project, a few minutes for a large one. Later calls re-analyze only what changed, so the map always matches the working tree.

Maps stay in your user cache (~/Library/Caches/layermap, ~/.cache/layermap or LAYERMAP_CACHE), never in the repository. LayerMap runs entirely on your machine and sends nothing anywhere (privacy, security).

ToolUse
project_explore_mapA directory, a file, or a declaration's callers and callees (direction INCOMING or OUTGOING, depth up to 8).
project_search_mapFind declarations by words in their names, paths or documentation.
project_find_referencesEvery usage of a declaration, compiled from current source.

The same from the command line:

npx layermap explore src/api/users.ts --name createUser --direction INCOMING --depth 8npx layermap search "invoice total"npx layermap refs src/billing/tax.ts calculateTax

FAQ

How long does the first map take? On a recent Mac:

  • Miniflux (400 Go files): about 3 s.
  • Polar's server (1,900 Python files): about a minute.
  • Conductor (1,500 Java and 1,300 TypeScript files): under two minutes.

What can't it see? Calls the compiler cannot resolve statically: dependency injection, unknown framework routing, reflection and computed names. The tools say where a trace stops, and a missing relationship does not prove absence.

Windows? Not yet.

How do I remove it?

  • Claude Code: /plugin uninstall layermap@layermap.
  • Codex: codex plugin remove layermap@layermap.
  • Then delete the cache directory.

Development

You need pnpm, Go 1.24 and a JDK 21+. A missing toolchain leaves its language unanalyzed.

Run pnpm install. On the first install pnpm asks which dependencies may run build scripts; allow esbuild only, since the SQLite package ships prebuilt binaries:

pnpm approve-builds esbuild '!@photostructure/sqlite'

Then run pnpm test, pnpm typecheck or pnpm lint.

pnpm package builds the npm package.

License

Apache-2.0. See LICENSE and NOTICE.

來源:README.md,提交 82b8dcf

工具

0
工具後設資料尚未被收錄。

版本歷史

1
  1. v0.1.4最新Oct 5, 2026