fakepose-mcp
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": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| list_presetsC | The preset families (pistol, rifle ...), each hold in them (side:mark[:kind]) with its source, and the captured grips (a hand-to-gun pose kept from an approved hold). |
| show_presetA | One preset in full: where the hand sits on the mark (palm facing, finger direction, offsets, push out) and each digit's rule (keep, or curl to contact toward the fist). |
| set_presetA | Adds or replaces one authored hold. The next run_hands uses it. |
| run_handsB | Runs Weapon Lab's Gun Kit hands step with fakepose (Blender, at most 10 minutes): the gun at aim, the holds from the presets, every clip keyed and written, the eye's clipping checked. Returns the summary. |
| hold_reportB | The last run's holds: each preset's source, the fingers' curls and what stopped them, the arm's angles (reported, not enforced), how far a hand was set down off the surface. |
| clip_reportB | The last run's clips: frames, the eye's clipping per clip (frames checked, frames with clipping, the most pixels, the deepest seen), and the first-person pictures (start, middle, end). |
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
Most tools are clearly distinct: list/show/set form an obvious preset CRUD subset, run_hands executes, and hold_report vs clip_report cleanly split hold vs clip diagnostics. Minor overlap exists because list_presets also surfaces captured grips while hold_report reports on holds, which could momentarily blur when to use which for grip data.
All names use snake_case and are predictable. The verb_noun pattern holds for list_presets, show_preset, set_preset, and run_hands, but the two report tools use noun_report instead of a verb prefix, a minor deviation.
Six tools is well-scoped for a focused Blender gun-kit hands pipeline: preset listing/inspection/authoring, one execution tool, and two complementary post-run reports. Every tool has a clear role with no filler.
The surface covers the main lifecycle: inspect presets, author/replace a hold, execute the hands step, and review hold and clip outcomes. The notable gap is a delete/remove operation for presets or captured grips, plus reports only reflect the last run, but these are workable around.