gto-recipe-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
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| searchC | 模糊搜索。kind = item | recipe | machine | material。 |
| recipeC | 单条配方全文。id 可为完整 id、type/tail 或唯一的 tail。 |
| producersB | 产出该物品/流体的配方，按 EU/t 升序；默认隐藏打包/解包/宇宙模拟/回收类，产物>8 种的排后（include_all=True 关闭）。 |
| consumersB | 消耗该物品/流体的配方，按 EU/t 升序（不含不可消耗输入）；降噪规则同 producers。 |
| machines_forB | 能跑该配方类型（或该配方所属类型）的全部机器，多方块优先。 |
| machineC | 机器定义：配方类型、仓室能力（PartAbility）、tooltip。 |
| materialC | 材料：化学式、组分、flags、形态，以及会不会被自动分解（及对应分解配方）。 |
| traceB | 文本树展开上游(up=谁产它、它要什么)/下游(down=谁吃它、产出什么)，每层取 EU/t 最低的 per_level 条。 |
| balanceC | 物料平衡。runs=[{recipe_id, runs}] 或 [{"": runs}]；tier(数字或 EV/IV…) 叠加概率 boost（GTCEu 默认公式，未核实 GTO）。 |
| me_partsA | ME 相关机器部件清单（id、中文名、是否可放进多方块）。 |
| planB | 自动配平产线（全部为每秒速率）。feed={物品: 每秒数量}；steps=[配方id 或 {id, mode=use|make, share, drive, target, machine, parallel, voltage, runs, perfect}]。 use（默认）：按驱动输入可用量定次数，驱动缺省=第一个属于进料或 use 产物、且不是 make 目标的输入；make：按 target 净缺口补料（如氢+氟→氢氟酸）；runs：固定次数/秒。 输出净外部需求/净产出/进料剩余/内部循环、总 EU/t，以及每步 runs/s、机器、超频、所需台数。 |
| balance_vC | balance 加电压：三栏同 balance（概率 boost 按该电压），并给每条配方超频后 EU/t、时长与机器秒。runs 同 balance，条目可加 machine/parallel/perfect。 |
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 12 tools
Most tools target distinct queries (recipe, producers, consumers, machine, material, me_parts). Some conceptual overlap exists: trace (upstream/downstream tree) relates to producers/consumers (flat lists), and balance vs balance_v differ only by added voltage, though detailed descriptions clearly delineate them.
All names use consistent lowercase snake_case, which is readable and predictable. Minor deviations exist: singular/plural mix (machine vs machines_for), a _for suffix, and the balance/balance_v abbreviation, but no chaotic mixing of conventions.
12 tools sit comfortably in the ideal 3-15 range. Each tool exposes a distinct capability (search, lookup, relationship tracing, balancing, planning) that earns its place for a complex recipe-mod domain.
The surface covers the recipe lifecycle well: search, recipe detail, producers/consumers, machine and material definitions, tracing, material balancing, production-line planning, and ME parts. Coverage is broad for the domain; only niche gaps (e.g. dedicated item/ore lookups) remain, which search can partially handle.