MCP server for What's Up Docker (WUD) - provides tools to interact with WUD API for managing containers, registries, triggers, watchers, and store configuration
WUD MCP server has 17 tools with basic naming and descriptions, but significant gaps in schema completeness and parameter documentation. All tools have descriptions (10-50 chars, below the 50-200 char baseline for LLM optimization), but most lack detailed parameter descriptions and output schema documentation. Tool names follow verb_noun convention (get_, list_, watch_, run_, delete_) which is good, but descriptions are minimal and do not explain WHEN to use each tool or what the LLM should expect in responses. No tool annotations (readOnlyHint, destructiveHint) despite clear risk classifications. Parameter schemas are present but sparse, most lack detailed descriptions of expected formats, ranges, or constraints. Error handling is basic; no recovery guidance or actionable error messages visible in test code.
Delete a container from WUD by ID
Get WUD application info including name and version
Get details of a specific container by ID
List all triggers associated with a specific container
Get WUD logger configuration including log level
Get details of a specific registry by type and name
Get WUD store configuration including path and file
Descriptions are too short (10-50 chars) and lack LLM-optimized context. Baseline for A+ tools is 50-200 chars. Current descriptions do not explain WHEN to use each tool, what it returns, or how it differs from similar tools (e.g., get_container vs get_container_triggers).
Parameter descriptions are minimal or missing. E.g., 'id' parameter in get_container has description 'Container ID' but does not explain format (UUID? string? Docker container hash?), whether it is required, or what happens if not found.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | C | 60 | 2026-07-28+ | v2 |
Get details of a specific trigger by type and name
Get details of a specific watcher by type and name
List all containers currently watched by WUD
List all configured container registries
List all configured triggers
List all configured watchers
Manually run a specific trigger on a container
Run a trigger with optional simulated container data
Trigger a manual watch on all containers
Trigger a manual watch on a specific container
No output schema documentation visible. Tools return JSON but LLMs cannot infer response structure. E.g., list_containers returns '[]' in tests but no schema documents what fields each container object contains or what IDs/references are available for chaining.
No tool annotations (readOnlyHint, destructiveHint, idempotentHint) despite clear risk classifications. delete_container is DESTRUCTIVE but not annotated; watch_all_containers is WRITE but not marked. LLMs cannot infer safety properties without explicit hints.
Error handling is basic. Tests show success/failure but no recovery guidance. E.g., if delete_container fails, LLM has no hint whether to retry, ask user, or try a different approach. No actionable error messages visible.