agentleFS
vvs1offvvs1off/fabric-skills/skills/check/SKILL.md

check

Skill4 starsChanged 12 days agoMITView on GitHub

Audit the current Fabric mod project (fabric.mod.json, build setup, dependencies, mixins, client/server separation, secrets) and output a prioritized report.

  • Reads credentials

80 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use daily-arxiv, exp-eval and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • LambenthanSkill

    daily-arxiv

    每日从 arXiv 拉取新论文,过滤相关性,auto-ingest 高优先级论文,检测 SOTA 更新

    Also in 5 projects that use this

  • LambenthanSkill

    exp-eval

    实验判决门:Review LLM 独立评判实验结果 → 4 种判决路径 → 自动更新 claims confidence、ideas status、graph edges

    Also in 5 projects that use this

  • JuliusBrusseeSkill

    backprop

    Bug → spec protocol. When a bug is found or a test fails, trace the cause, decide whether a new §V invariant would catch recurrence, append to §B. This is the one non-obvious thing SDD does that plan-then-execute doesn't. Triggers on test failure, bug report, post-mortem, or explicit user ask.

    Also in 4 projects that use this

  • skyllwtSkill

    exp-pilot-eval

    Pilot result evaluation — read pilot results, apply success criteria, update idea page (pilot_result, failure_reason if failed), generate PILOT_VERDICT_REPORT. Called by /ideate Phase 5 after /exp-pilot-run.

    Also in 4 projects that use this

  • skyllwtSkill

    exp-pilot-run

    Pilot experiment execution — read Pilot Spec YAML, write pilot code, run experiment(Confirm with the user before operation and require the applicant to conduct manual inspection), return results. Called by /ideate Phase 5. Does NOT modify wiki pages or judge pass/fail.

    Also in 4 projects that use this

  • +6 more from Lambenthan
  • +1 more from skyllwt

The file

We split this skill into the rules it gives an agent

6 rules. 0 are common, 0 are distinctive, and 6 are very specific to this project.

  • 0

    Common

    Most of the 69 edited versions we read keep these.

  • 0

    Distinctive

    This author's own choices. Worth a look.

  • 6

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Use the checklists from fabric-project-setup, fabric-mixins, fabric-networking and fabric-release.

    line 13
  2. Very specific:

    Every entrypoint class exists at that FQCN and implements the right interface (ModInitializer, ClientModInitializer, …

    line 37
  3. Very specific:

    Every mixin config listed exists.

    line 43
  4. Very specific:

    Search common code for client-only references: grep -rnE …

    line 50
  5. Very specific:

    No secrets: grep for token, secret, password, api_key, MODRINTH, CURSEFORGE with values in tracked files.

    line 67
  6. Very specific:

    Run ./gradlew build (this writes only to build/) and include the result.

    line 74

“Common” is measured: kept by at least 6 in 10 of the 69 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

/fabric-skills:check - audit this Fabric mod

Audit the Fabric mod in $ARGUMENTS (or the current directory if empty). Read-only: do not modify files. Produce a report; offer fixes afterwards.

Use the checklists from fabric-project-setup, fabric-mixins, fabric-networking and fabric-release. For a deep mixin review, delegate to the mixin-reviewer agent.

What to inspect

1. Era and versions

  • gradle.properties, build.gradle, settings.gradle, wrapper version.
  • Era: 26.1+ (plugin net.fabricmc.fabric-loom, no mappings) vs older (remap plugin + Yarn/Mojang mappings). Flag mixtures (e.g. a mappings line on 26.1+, modImplementation with non-remapping Loom, implementation for mods with remapping Loom).
  • Compare versions with https://fabricmc.net/develop/ for that Minecraft version, if reachable; report outdated Loader/Fabric API/Loom (informational, not an error). When mentioning newer Minecraft versions, check whether they are releases or snapshots (Mojang's version_manifest_v2.json latest.release, or stable in meta.fabricmc.net); Fabric API Maven metadata also lists snapshot builds and must not be reported as a release.
  • -SNAPSHOT / + dynamic versions (e.g. loom_version=1.19-SNAPSHOT) make builds non-reproducible; suggest pinning for release builds.
  • Java release consistency: options.release, sourceCompatibility, mixin compatibilityLevel, fabric.mod.json "java".

2. fabric.mod.json

  • Valid JSON, schemaVersion: 1, id format, "version": "${version}" with processResources expansion.
  • id matches loom.mods, MOD_ID, assets/<id>/, mixin config names.
  • Every entrypoint class exists at that FQCN and implements the right interface (ModInitializer, ClientModInitializer, DedicatedServerModInitializer, ModMenuApi).
  • environment matches reality (client-only mod with server code, or vice versa).
  • depends: minecraft range not "*" and matches minecraft_version; fabricloader, java, fabric-api present when used. Optional integrations in suggests/recommends, not depends.
  • Every mixin config listed exists; client configs have "environment": "client".
  • The icon path (if set) exists under src/*/resources.
  • accessWidener file exists and is also set in loom.accessWidenerPath.

3. Client/server separation

  • Is splitEnvironmentSourceSets() enabled? If not, flag as high severity for non-client-only mods.
  • Search common code for client-only references: grep -rnE "net\.minecraft\.client\.|com\.mojang\.blaze3d|ClientPlayNetworking|MinecraftClient|KeyMapping|KeyBinding" src/main
  • Client mixin targets (net.minecraft.client.*) listed in a common mixin config.
  • Payload types registered only in a client initializer.

4. Mixins

  • List every mixin class, its target and injector types.
  • Flag: @Overwrite, @Redirect, cancellable HEAD injections on hot methods, ordinal, missing @Unique/mod-id prefixes, defaultRequire: 0 or require = 0, mixins that could be a Fabric API event, non-mixin classes in mixin packages.
  • For more than a few mixins, run the mixin-reviewer agent and include its findings.

5. Networking (if present)

  • Payloads use the API generation matching the version; types registered in common init.
  • Server receivers validate input (ranges, distance, entity existence, permissions).

6. Security and hygiene

  • No secrets: grep for token, secret, password, api_key, MODRINTH, CURSEFORGE with values in tracked files; no .env, keystores, *.key.
  • .gitignore covers build/ .gradle/ run/ out/ .idea/ logs/ crash-reports/.
  • Gradle wrapper present and committed; gradlew has LF line endings.
  • Release workflow (if any) uses ${{ secrets.* }} only, minimal permissions, pinned actions.

7. Build

  • Run ./gradlew build (this writes only to build/) and include the result. If it fails, put the first error in the report.

Report format

# Fabric mod audit - <mod name> (<mod id>)
Minecraft <v> | Loader <v> | Fabric API <v> | Loom <v> (<plugin id>) | Java <n> | Mappings <...>
Build: PASS / FAIL (<first error>)

## Critical (will crash or break)
- [file:line] problem -> fix

## Warnings (fragile, incompatible, or wrong-era)
- ...

## Suggestions
- ...

## Looks good
- short list of things checked and fine

Rank by severity; cite file:line for every finding; give a concrete fix for each. End by asking which findings the user wants fixed.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.