CRAP Score MCP
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| CRAP_BASE | No | Starting Git revision used for reviewing a longer change. Optional; CI uses the push or PR base when unset. | |
| CRAP_PORT | No | Overrides the foreground MCP service port. The local dashboard defaults to http://127.0.0.1:47731/. | |
| CRAP_DATA_DIR | No | Overrides the foreground service data directory. The machine installer stores its private runtime under ~/.local/share/crap-mcp by default. |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| analyze_function_healthC | Inspect a symbol in an enabled language by repository-relative path and name or line. Returns measured evidence and matching findings. |
| scan_repository_hotspotsC | Rank and page findings. Set refresh for background scan progress; use the snapshot id as a task baseline. |
| get_repository_reportC | Page through all assessed functions, including acceptable and unscored ones. Pass the first snapshot ID on subsequent pages for a consistent report. |
| review_changesB | Review a session against its starting snapshot or revision. Use format=chat at completion and show the returned session health and raw CRAP changes in your reply. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 4 tools
The tools target distinct levels of the domain: single-function analysis, repository hotspot ranking, full repository reporting, and change-session review. There is some conceptual overlap between hotspot findings and the full report, but the descriptions make the scope difference clear enough to avoid most misselection.
All tool names use snake_case with a clear verb_noun pattern: analyze_function_health, scan_repository_hotspots, get_repository_report, review_changes. The conventions are predictable and consistent throughout.
Four tools are well-scoped for a CRAP score/code-health server. Each tool maps to a distinct workflow step, and there are no redundant or filler tools.
The surface covers core workflows: scanning repository hotspots, reporting assessed functions, inspecting a specific function, and reviewing changes against a baseline. A few secondary operations, such as explicit scan-status or repository discovery, are not separate tools, but the main CRAP-score lifecycle is covered.