Use before grep or reading files when a task asks what calls a function or method, what changing it affects (up to HTTP routes, handlers, jobs and commands), where code is used, or how the code is organized. The layermap MCP tools answer these from a static map of the repository in one call.
Installs into .claude/skills of the current project.
Are you the author of Layermap?
Add the live security badge to your README. It updates with every re-scan.
[](https://www.skillsdirectory.com/skills/coffeecoproject-layermap)
---
name: layermap
description: Use before grep or reading files when a task asks what calls a function or method, what changing it affects (up to HTTP routes, handlers, jobs and commands), where code is used, or how the code is organized. The layermap MCP tools answer these from a static map of the repository in one call.
---
# LayerMap
The layermap MCP server is available here. For questions about what calls a function, what a change affects or where code is used, start with its project_explore_map tool (direction INCOMING, depth up to 8) before grep, then confirm in source.
- `project_explore_map` with `path` "." shows what the repository contains; a file `path` lists its declarations; with `name` (and `line` when names repeat) it shows a declaration's callers and callees. `direction` INCOMING with `depth` up to 8 traces callers up to the entry points; continue from any NOT EXPANDED declarations it names.
- `project_search_map` finds a declaration by words in its name, path or documentation.
- `project_find_references` lists every usage of a declaration, compiled from current source.
The map is static: dynamic dispatch, framework wiring and unresolved targets need source reading, and a missing relationship does not prove absence. A large repository's first map takes minutes to build; until then the tools say so, and you can read source meanwhile.
