Fabric Mod Development Agent Skill

Develop, inspect, scaffold, build, and diagnose Fabric 1.21.4 mods with ModCrafting's local MCP tools.

Context cost
1.3k tokensestimated from the bundle, loaded when it triggers
Bundle
1 file · 5.1 kBtext throughout, nothing executable
Licence
not declarednone named by its author
Last change
—no release on file
Servers it uses
Noneruns standalone

What it does

Develop, inspect, scaffold, build, and diagnose Fabric 1.21.4 mods with ModCrafting's local MCP tools.

What this skill changes about the agent has not been set out here as a list. Its SKILL.md, shown below, says it in the author’s own words.

gamedevmcp

The skill itself

A skill is instructions the model reads, so you can see what you are getting: below is the start of its SKILL.md. The front matter loads with every session, the body loads when the skill triggers, and the whole file comes with the install.

SKILL.md5.1 kB · preview
--- name: fabric-mod-development ---
6Use this skill for a Fabric 1.21.4 mod project on Windows.
7
8For all user-facing guidance, follow modcrafting-start: explain outcomes in the user's language, keep tool names and protocol details internal, and continue the requested work. For onboarding or visual creation, call modcrafting_open_workbench with the current project (content/recipe/gui as appropriate); merely returning an environment browserUrl does not open a browser. Never claim it opened unless the opener reports success, and even then say the system was asked to open it.
9
101. Start with modcrafting_environment_status and fabric_project_inspect.
112. Before registering a block, item, entity, enchantment, or recipe, call minecraft_data_lookup for every vanilla ID and property used. Do not invent IDs or hardness, resistance, stack-size, or enchantment data from memory.
123. Use fabric_docs_search for Fabric APIs and mc_wiki_search for game concepts.
134. Begin each chat task with modcrafting_session({action:"create",projectPath}). Make narrow edits with the host Agent's normal file tools, then call fabric_validate and start fabric_build_start({sessionId,projectPath,task:"build"}). Pass that same sessionId to status and stop calls; only resume it explicitly when continuing an older task.
145. Poll with fabric_job_status; interpret failures from Gradle output before attempting another edit.
15
16For new content, use fabric_content_workspace(action="template", kind=...) to obtain supported defaults instead of guessing parameters; then customize, preview and save. The template is data only, does not write files and does not replace required vanilla-data lookup.
17
18The plugin does not install a JDK. The environment check and builds resolve the same JDK 21: explicit JAVA_HOME first, otherwise PATH; the resolved home is passed to Gradle. MCP scaffolding includes the pinned Windows Gradle Wrapper; existing projects must provide gradlew.bat. First builds may download Gradle and dependencies. Report actual storage paths; do not move/delete old caches or hardcode machine paths.
19
20For visual recipes, obtain browserUrl from modcrafting_environment_status when using --web. The browser and fabric_recipe_workspace share the same project files. Use action catalog/list/read/save. Before updating, read and pass the exact revision as expectedRevision; use null only for a new ID. On RECIPE_CONFLICT, read and compare instead of blindly overwriting. Five vanilla serializers are editable: shaped, shapeless, smelting, blasting, stonecutting. Unsupported serializers or ingredient forms remain read-only. Saving requires rebuild/restart or explicit game reload before testing. Avoid ambiguous inputs matching vanilla recipes and verify the actual output in game.
21
22For visual content parameters, use fabric_content_workspace list/read/preview/save and the same browser project. Supported declarative types are item, food, block, entity, tool and armor. Read the document; preview with its expectedRevision (null for a new ID), review changed files, then save with the returned previewRevision and the same expectedRevision. A conflict requires read/compare; never bypass it. Handwritten or externally modified generated Java is read-only. Custom textures must reference an existing project PNG; vanilla previews use an optional local cache and may show a labeled type illustration. Entities reuse pig/zombie appearance and AI. Sword mining rules and armor appearance follow vanilla types/materials. Rebuild and restart before validating Java changes. Observer capability revision 5 exposes food/block/entity parameters and integrated-server equipment attributes; client attack/knockback attributes may lag. Validate actual effects and cleanup, not only command dispatch or generated source.
23
24For MC UI creation, use fabric_gui_workspace template/list/read/preview/save and the MC interface editor in browserUrl. template returns a data-only layout and supported component defaults; it does not write files. Supported layouts are custom-screen, option-list and hud-overlay. Use stable unique IDs, GUI-pixel coordinates, native button/toggle/slider controls and labels/panels/display slots. Labels may bind to local control values or player health/food. Buttons only close, reset defaults or increment a chosen slider; no arbitrary scripts or commands. Display slots are not server containers. Read the current revision; preview generated files with expectedRevision, then save with both expectedRevision and previewRevision. Conflicts preserve drafts; external Java is read-only. Rebuild/restart, press the configured key in game, validate actual widget messages/geometry and persisted configuration. Screen closes via its close button or normal Esc; HUD key toggles visibility. Browser interaction only simulates state. HUD trace assertions remain INCONCLUSIVE in the existing engine; report fresh native draw observations/screenshots separately, never as an automatic HUD PASS.
This is a preview of SKILL.md. Download or install the skill to get the whole thing.
In the file
SKILL.md674 words
Files1
Licencenot declared
Why you can read it

Nothing in a skill executes. The client loads the text and the model follows it, so a skill can be audited the way a runbook is — by reading it.

What it costs in context

Skills are not billed by the call. They are paid for in context: every token the instructions occupy is a token your code, your diff and your conversation cannot use. Here is what this one takes and when it takes it.

≈40
always loaded
The name and description, so the model knows the skill exists and when to reach for it.
1,235
on trigger
The instruction body, read only when the skill fires.
0.64%
of a 200k window
Ten skills this size would take about 6% of the window before you open a file.
050k100k150k200k context window

1.3k tokens, estimated from the bundle at four bytes to the token, held for the rest of the session once it triggers. Middling. Fine to keep on in a project where you use it weekly, worth unloading in one where you never do.

Servers bill, skills cost

A server charges by the month. A skill charges once per session, in context, and then keeps charging it for as long as the session lives.

Before and after

The same question, put to the same model twice: once as it comes, and once with these instructions loaded.

No worked example has been published for this skill yet.

The procedure it runs

The procedure has not been set out here step by step. It is in the body of its SKILL.md: the start of it is shown above, and the whole file comes with the install.

Prose, not code

These steps are written for a model to follow, not executed by a runtime. It can still be told to skip one, and it will say so when it does.

Servers it uses

None. This skill calls no MCP servers at all.

Everything it needs is in the instructions, so it works in a project with nothing connected — the model reads the file and changes how it works with what it can already reach.

Its SKILL.md does not declare which tools it asks the agent for, so what it touches is whatever the instructions themselves ask for — read them before you install it.

What it asks for
Writes filesnot declared
Network accessnot declared

This SKILL.md carries no allowed-tools line, so neither is declared. A skill grants no permissions of its own — it can only ask for tools your client already has.

What it will not do

Every skill is narrow, and the useful ones say where they stop. These are the jobs this one is the wrong tool for.

What this skill is not for has not been published here. Nothing is implied by that: it is a section the author has not filled in.

What is in the bundle

1 file, 5.1 kB on disk. A bundle is text throughout: the instructions the model reads, plus the templates it fills in.

  • SKILL.md5.1 kB
What is not in it

No dependencies and nothing executable: a skill is text the agent reads, so the bundle is 1 file you can review in full before installing.

Install

Installing copies the bundle into your project. Nothing runs at install time — the files sit on disk until the model reads them.

# Fabric Mod Development · 1.3k tokens when loaded · macOS, Linux mkdir -p .claude/skills && t=$(mktemp -d) && curl -fsSL -o "$t/bundle.tar.gz" https://mcprush.com/api/skills/fabric-mod-development/bundle.tar.gz && tar -xzf "$t/bundle.tar.gz" -C "$t" && [ -d "$t/fabric-mod-development" ] && { [ ! -e .claude/skills/fabric-mod-development ] || { mv .claude/skills/fabric-mod-development "$t/.previous" && echo "Previous copy moved to $t/.previous"; }; } && mv "$t/fabric-mod-development" .claude/skills/ && echo "Installed .claude/skills/fabric-mod-development"
On Windows, or any OS with Node 20+
npx mcprush@latest skill add newstarbar/fabric-mod-development

Windows PowerShell says 'running scripts is disabled on this system'? Run Set-ExecutionPolicy -Scope CurrentUser RemoteSigned once, or type npx.cmd instead of npx (codex.cmd, gemini.cmd, grok.cmd likewise). PowerShell 7 needs neither. Writes to .claude/skills/fabric-mod-development/ in the current project. For every project instead, run mkdir -p ~/.claude/skills && t=$(mktemp -d) && curl -fsSL -o "$t/bundle.tar.gz" https://mcprush.com/api/skills/fabric-mod-development/bundle.tar.gz && tar -xzf "$t/bundle.tar.gz" -C "$t" && [ -d "$t/fabric-mod-development" ] && { [ ! -e ~/.claude/skills/fabric-mod-development ] || { mv ~/.claude/skills/fabric-mod-development "$t/.previous" && echo "Previous copy moved to $t/.previous"; }; } && mv "$t/fabric-mod-development" ~/.claude/skills/ && echo "Installed ~/.claude/skills/fabric-mod-development" (macOS, Linux) — or add --global to the npx line. The curl line replaces .claude/skills/fabric-mod-development/ only after the download succeeds, so files left there by another skill of the same name, or by an older release, do not mix in. A copy already there is moved aside, not deleted, and the line prints where.

Which clients pick it up on their own

A skill is a folder of text. Most of these read a skills folder and pick it up without being told; the rest take the same folder another way — an upload in the app, a parameter on an API call — and then choose it by its description just the same.

Claude Code.claude/skills/
Claude DesktopCustomize › Skills
ChatGPTSkills › Create › Upload
Cursor.cursor/skills/
VS Code.github/skills/
Codex CLI.agents/skills/
Gemini CLI.gemini/skills/
Grok Build.grok/skills/
Microsoft CopilotSkills › Upload a skill
PerplexityComputer › Skills › Upload a skill
DeepSeek Harness.dsh/skills/
Zed.agents/skills/
Devin Desktop.devin/skills/
Windsurf (legacy).windsurf/skills/
OpenAI Agents SDKInstructions or sandbox Skills
HTTP / APIa request parameter
This release
Versionnot versioned
Publishedno release date on file
PriceFree
Referencenewstarbar/fabric-mod-development

Versions

Its author publishes no version number, so there is nothing here to pin to: what you install is the folder as it stands today. Instructions change more often than APIs do — a skill can be rewritten entirely without anything it depends on moving.

Pinning

Nothing to pin to: this skill carries no version number of its own. What you install is what the folder holds on the day you install it.

Reviews

—no reviews yet

Nobody has reviewed this skill yet. The rating is the mean of the reviews written here, so there is none until somebody writes the first.

Sign in to review it

A review is tied to an account so it can be answered, edited and taken down by the person who wrote it.

Sign in
Who can post

Anybody with an account, one review each. It carries the version that was current when you wrote it, and the publisher may reply once.

Who wrote it

NE
newstarbar

Publishes on mcprush.

0 servers listed1 skill listednot claimed
Profile
Publisher
Servers0
Claim this skill