character-forge-mcp

Overview Schema Source Related Servers Discussions Admin

Generate T-pose concept sheet

character_concept

Generates a front-and-back concept model sheet from a saved character card, and can apply a reference image's pose; returns the prompt only when image generation is unavailable.

Instructions

Builds an A-pose model-sheet prompt from a saved card's look and, with FAL_KEY set, generates it on fal.ai: a faceless figure built from look.body (the character's own proportions) is drawn front and back and given to GPT Image as the pose reference, with the full look → concept.png (the card is re-rendered with it). pose_image puts the character in the pose of any picture → concept-pose.png. Without FAL_KEY (or with dry_run) returns the prompt only. Costs ≈ $0.02 per sheet.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "id": "<id>"
    },
    "name": "character_concept"
  }
}

Input Schema

Name	Required	Description
`id`	Yes	Character id returned by character_save.
`mode`	No	How the figure steers the image. gpt (default, GPT Image edit ≈ $0.02): the figure is a pose reference. img2img (Z-Image ≈ $0.01): it is the starting canvas; greys colours. pose/depth/canny: ControlNet conditions (weak on this endpoint). Kept for experiments; leave unset.
`seed`	No	Seed, to reproduce or vary a result.
`extra`	No	Extra art direction appended to the prompt.
`style`	No	Omit for the house style (one look across the library). 'card' = the designer's suggestion in look.art_style; or any style text, e.g. 'anime cel-shaded'. Non-default styles are saved as concept-<style>.png next to concept.png.
`views`	No	Views on the sheet. Default front-back.
`figure`	No	The mannequin: blocking-lit (default with look.blocking: zone colours with volume), blocking (flat 2D colour patches), shaded (grey 3D), flat, mask.
`dry_run`	No	Only return the prompt, do not call the image API.
`strength`	No	img2img: how much to repaint the mannequin (default 0.85 ≈ 7 of 8 steps; lower keeps the pose harder but greys the colours, higher frees colours but loses the pose).
`pose_image`	No	Path, URL or data URI of a picture whose pose the character should take (e.g. their idle pose or signature gesture).
`control_scale`	No	ControlNet modes: how strictly to follow the condition (default 0.75).

Schema Changelog

Changes observed during successful MCP inspections.

First observedv0.3.1Nov 12, 2026

TDQS

A3.9/5.0

Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations, the description carries the full burden and does well: it discloses the FAL_KEY requirement, approximate cost (~$0.02 per sheet), that dry_run skips the API call, and that non-default styles are written as concept-<style>.png alongside concept.png. The one soft spot is '(the card is re-rendered with it)', which leaves unclear whether the saved card is persistently mutated or only the local file is updated.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Front-loaded with the primary action and the key conditional (FAL_KEY), then layers output names, the pose_image branch, and cost. It is a single dense run-on paragraph with arrow notation, which is efficient but harder to scan than separated sentences; no sentence is wasted, though.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For an 11-parameter generative tool with no annotations and no output schema, the description covers cost, auth, fallback behavior, output filenames, and the card-re-render side effect. Gaps are the ambiguous persistence of the card update and no mention of failure behavior when the API call fails; the title/description pose mismatch (T-pose vs A-pose) is also left unresolved.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100% and the schema descriptions for mode, style, figure, and strength are already detailed, so the description's parameter content mostly restates it. It does add one genuinely new fact — pose_image produces concept-pose.png — and reinforces that pose_image accepts a picture of any pose. Baseline-3 is appropriate since the schema does the heavy lifting.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description gives a specific verb and artifact: it builds an A-pose model-sheet prompt from a saved card's look and, with FAL_KEY, renders it on fal.ai into concept.png or concept-pose.png. That is far more informative than the title, which says 'T-pose' and mismatches the body's 'A-pose'. It implicitly separates itself from prompt-only siblings by naming the generated files, but does not name a sibling explicitly.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

It states the conditions that change behavior clearly: without FAL_KEY or with dry_run it returns the prompt only, and pose_image switches the tool to reproducing a reference picture's pose. It also advises leaving the ControlNet modes unset ('Kept for experiments'), which is actionable. It stops short of saying when to prefer a sibling tool over this one.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.