kuma
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
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 |
|---|---|
| kuma_contextA | Context & memory recall. Call FIRST each session. Lean — each call returns only what you need. CORE ACTIONS (use these): |
| kuma_kuma_contextA | Context & memory recall. Call FIRST each session. Lean — each call returns only what you need. CORE ACTIONS (use these): |
| kuma_memoryB | Persistent knowledge. Record what matters, skip what doesn't. This is what saves time in future sessions. Don't record what grep/glob answers faster (functions, imports, types, components). CORE ACTIONS (use these): |
| kuma_kuma_memoryA | Persistent knowledge. Record what matters, skip what doesn't. This is what saves time in future sessions. Don't record what grep/glob answers faster (functions, imports, types, components). CORE ACTIONS (use these): |
| kuma_safetyA | Safety & verification. Use at task boundaries, not on every edit. CORE ACTIONS (use these): |
| kuma_kuma_safetyA | Safety & verification. Use at task boundaries, not on every edit. CORE ACTIONS (use these): |
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 6 tools
The three functional groups (context, memory, safety) are reasonably distinct from each other, but each one is duplicated as an identical twin with a byte-for-byte identical description (kuma_context vs kuma_kuma_context, etc.). An agent has no basis to choose between a tool and its clone, guaranteeing arbitrary misselection in half the available surface.
Names follow a readable kuma_verb/noun snake_case pattern, but the redundant 'kuma_kuma_' prefix on three tools breaks the convention and looks like an accidental double-registration rather than a deliberate scheme. Consistent enough to parse, inconsistent enough to mislead.
Six tools would be a fine scope for this workflow, but three of them are exact duplicates, leaving only three effective tools covering context, memory, and safety. The apparent count overstates the real surface, though three genuine umbrellas is acceptable for the stated purpose.
Each umbrella tool exposes a rich action set (init/research/history/map/impact, gotcha/decision/search, guard/verify/checkpoint/rollback), covering the session lifecycle from recall through record to verification and rollback. Coverage looks solid; the main gap is that the duplicate twins add no capability, only noise.