New: 98,000+ skills indexed from GitHub repos with 100+ stars · every skill safety-checkedBrowse top skillsTop skills
Esc to closeBrowse all skills
Submit a skillBrowse skills
Submit a skillBrowse skills

Agent skill

Anime Tools

by sorryhyun in sorryhyun/anima_lora

The trainer ↔ animetools boundary — what the curation split moved out, the typed request/stage API the make targets build, the git-pin dev loop and its stale-venv trap, and the tests that guard the…

MITAuto-check passedDevelopment

Install Anime Tools

skills CLI
$ npx skills add sorryhyun/anima_lora --skill anime-tools -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install sorryhyun/anima_lora anime-tools --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/sorryhyun/anima_lora.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.claude/skills/anime-tools .claude/skills/anime-tools && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
anime-tools
GitHub stars
125
Token cost
~1.5k tokens
SKILL.md length
695 words
Files
1
Skills in repo
12
Repo updated
First seen
Licence
MIT

At a glance

The trainer ↔ animetools boundary — what the curation split moved out, the typed request/stage API the make targets build, the git-pin dev loop and its stale-venv trap, and the tests that guard the…

  • Development work in your project
  • SKILL.md covers What lives in the package, The typed request API is the…, The pin and Moving more code into the…
  • Calls uv, make and python

What it does

Anime Tools is an agent skill from sorryhyun/anima_lora. The trainer ↔ animetools boundary — what the curation split moved out, the typed request/stage API the make targets build, the git-pin dev loop and its stale-venv trap, and the tests that guard the contract. Load before importing animetools, editing scripts/tasks/, adding or changing a stage, bumping the pinned tag, or debugging a stage's argv.

Its SKILL.md is about 1.5k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in Development. It works with Git. The repository describes itself as: optimized anima lora training script. The licence is MIT.

When your agent uses it

  • Development work in your project

Example prompts

  • “/anime-tools”

Requirements

  • Python 3

What it can do on your machine

Read from SKILL.md and the folder at commit d16b651. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • uv
    • make
    • python

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    Links to these hosts (documentation or services it may open):

    • github.com

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Anime Tools loads about 1.5k tokens when it runs. Until then it costs about 90 tokens; SKILL.md has 695 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~90
When it runs · the whole SKILL.md, loaded when a task matches
~1.5k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from sorryhyun/anima_lora at commit d16b651, republished under its MIT licence (© sorryhyun). 695 words, ~1,484 tokens.

Download SKILL.mdSave it as .claude/skills/anime-tools/SKILL.md (or your agent's skills folder).
name
anime-tools
description
The trainer ↔ anime_tools boundary — what the curation split moved out, the typed request/stage API the make targets build, the git-pin dev loop and its stale-venv trap, and the tests that guard the contract. Load before importing anime_tools, editing scripts/tasks/, adding or changing a stage, bumping the pinned tag, or debugging a stage's argv.

Curation lives in anime_tools

Sibling repo https://github.com/sorryhyun/anime_tools (package anime_tools, checkout ../anime_tools). Per-feature contracts: ../anime_tools/docs/. The package carries its own skills — read them before editing package code: ../anime_tools/.claude/skills/{captions,add-stage,model-catalog,release}/SKILL.md.

What lives in the package

Caption grammar, tag taxonomy/correction, variants sidecars, caption index, the Anima Tagger, the caption-master stages (autotag / position clauses / multiview audit), masking (SAM3 / merge), grouping (PE-Spatial near-twin features → groups.json), and the tagger-only benches + their gitignored training history. Import anime_tools directly — the library._moved shims and every forwarding shell are gone.

Dependency direction is trainer → anime_tools, never the reverse. tests/test_curation_boundary.py guards the trainer side; the package guards itself.

What stays trainer-side:

  • GUI panels — they reach the package only via the autotag_server stdio protocol, daemon jobs, and the torch-free grammar.
  • library/models/pe.py — a permanent re-export. The vendored PE vision tower is owned by anime_tools.vision.pe, so a standalone anime_tools groups with the same PE-Spatial-B16-512 the trainer uses. The library/vision/ encoder/bucket registry stays here.
  • configs/clause_vocabulary.yaml — the user-editable override of the package default.
  • Benches that use the tagger as a judge of DiT output.

The typed request API is the front door

One frozen request dataclass per stage (ResizeRequest, AutotagRequest, PositionRequest, CorrectRequest, GroupRequest, SamMaskRequest, …), registered in anime_tools.stages.registry with a lazy Stage.runner(). python -m anime_tools.<pkg>.cli.<name> is its shell — use it directly for the tagger CLIs that have no request yet (make daemon-run ARGS="-m anime_tools.tagger.cli.train_sidecar …").

The make targets (preprocess-resize, mask, curate-group, caption-*, tagger, autotag) keep their names and --queue routing. The rule for the wrappers in scripts/tasks/: build a request, never spell a flag. _common.execute_stage runs it

  • in-process under a daemon job — one interpreter, one tagger/SAM3 load shared across consecutive stages, released before a trainer child via release_models(); or
  • as a python -m <stage.module> *req.to_argv() child from a shell.

A user's ARGS reach a stage through the request's own generated parser (request_with_args), so every flag the stage has still works from make and an unknown one fails with the stage's usage. run() and the in-process path both export ANIMA_HOME so the package's bare relative defaults anchor on this checkout (ANIME_TOOLS_HOME → ANIMA_HOME → CWD).

Guard: tests/test_anime_tools_cli_contract.py re-parses every emitted argv through the stage's parser and is the drift alarm. The release tag is the version: a surface change shows up as a failing contract row (or a TypeError on a request field) when the pin moves.

Adding a stage or a flag? Follow ../anime_tools/.claude/skills/add-stage/SKILL.md, then add the trainer-side wrapper + a contract-test row here.

Show full SKILL.md (303 more words)Show less

The pin

A git dependency, not PyPI: pyproject.toml pins a release tag (tag = "vX.Y.Z") under [tool.uv.sources] (a plain project dependency) — cut with the package's release skill (version bump → annotated tag → release.yml).

The trainer .venv holds the pinned copy, not ../anime_tools. An edit in the sibling checkout is invisible to make targets, daemon jobs and the GUI until a tag is cut and the pin moves — this has silently run stale package code on the GPU before. python -c "import anime_tools" with cwd=../anime_tools lies (sys.path[0]); check from the trainer root.

  • Ship a package change: release it upstream (tag pushed), move the tag in pyproject.toml, then uv lock --upgrade-package anime-tools && uv sync.
  • The package's [tool.uv.sources] leak into this lock. uv honors a git dependency's own sources, so a torch index pinned upstream collides with the trainer's rocm-windows group at uv lock. Any upstream torch source must be extra/group-conditioned so a consumer never sees it.
  • Testing an unreleased change: PYTHONPATH=../anime_tools (it shadows site-packages). There is deliberately no path-source dev group — a ../anime_tools path source lands in uv.lock, and then a flagless uv sync (the updater, install.sh / install.ps1) fails on every machine without the sibling checkout.
  • Smoke an unpushed change on the GPU: submit with DaemonClient.submit_command(argv=["-m", ...], extra_env={"PYTHONPATH": "/home/sorryhyun/anima/anime_tools"}) — PYTHONPATH matches none of the daemon's CAPTURED_ENV_PREFIXES, so make daemon-run cannot pass it.
  • uv sync also prunes ad-hoc-installed packages — reinstall anything you added by hand after a sync.

Moving more code into the package

  • The package's ruff rules are stricter than the trainer's (SIM115, DTZ005, EXE001…) — expect to fix those on any move.
  • tests/test_doc_refs.py treats any slash-path token in a tracked .md as a live path. Mention deleted files in dotted-module form, not as paths.
  • ../anime_tools/bench/_common.py is a copy of the trainer's envelope helper; a probe there that reads the trainer's results dir must be run from this checkout.

© sorryhyun, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in .claude/skills/anime-tools of sorryhyun/anima_lora.

Open the folder on GitHubat commit d16b651

Compare with similar skills

Anime Tools next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Anime Tools compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Anime Tools this skillsorryhyun/anima_lora125—~1.5kAutomated safety check: PassMIT
Finishing a Development Branchobra/superpowers297k5 repos~1.9kAutomated safety check: PassMIT
Code Review ChecklistshareAI-lab/learn-claude-code78k4 repos~1.1kAutomated safety check: PassMIT
Code Design Rationale Investigatorcursor/plugins11k9 repos~2.6kAutomated safety check: PassNone
Contributor-First PR MergeHKUDS/OpenHarness16k1 repos~847Automated safety check: PassMIT
Finishing A Development Branchfarm-fe/farm5.6k34 repos~1.8kAutomated safety check: PassMIT

Similar skills

  • Finishing a Development Branch

    obra/superpowers

    Walks the last step of a branch: confirm tests pass, detect the git environment, ask how to integrate, carry out your choice and clean up the worktree.

    297k GitHub starsUsed in 5 repos~1.9k tokens
    DevelopmentAuto-check passed

  • Code Review Checklist

    shareAI-lab/learn-claude-code

    Reviews code against a five-part checklist covering security, correctness, performance, maintainability and testing, and reports findings in a fixed format.

    78k GitHub starsUsed in 4 repos~1.1k tokens
    DevelopmentAuto-check passed

  • Code Design Rationale Investigator

    cursor/plugins

    Official

    Digs into why code is shaped the way it is by checking git history, pull requests and connected tools in parallel, then reporting a cited read on the tradeoffs.

    11k GitHub starsUsed in 9 repos~2.6k tokens
    DevelopmentAuto-check passed

  • Contributor-First PR Merge

    HKUDS/OpenHarness

    Merges external GitHub pull requests while keeping the original author credited, and fixes conflicts after the merge instead of rewriting the contribution.

    16k GitHub starsUsed in 1 repo~847 tokens
    DevelopmentAuto-check passed

  • Finishing A Development Branch

    farm-fe/farm

    A skill your agent uses when implementation is complete, all tests pass, and you need to decide how to integrate the work - guides completion of development work by presenting structured options for…

    5.6k GitHub starsUsed in 34 repos~1.8k tokens
    DevelopmentAuto-check passed

  • Migrate Internal Package into Ghost

    TryGhost/Ghost

    Moves a package from another TryGhost repository into Ghost as an internal workspace package while keeping its Git history, with checkpoints for the steps that need an administrator.

    56k GitHub stars~3.8k tokensUpdated today
    DevelopmentAuto-check passed

More from sorryhyun/anima_lora

All 12 skills in this repo

  • Model Catalog

    sorryhyun/anima_lora

    The model catalog (library/downloads.py) — one Asset row per weight (repo, files, destination, installed probe), packs, resolve() name order, and the rule that loaders import their default paths…

    125 GitHub stars~502 tokensUpdated today
    Auto-check passed

  • Qwen21

    sorryhyun/anima_lora

    Qwen-Image-2.1 LoRA line (NOT Anima) — running cache/train through the daemon, make gui-qwen, the CacheRequest/TrainRequest flag surface and how to add a field, model-dir resolution, cache layout…

    125 GitHub stars~1.9k tokensUpdated today
    Auto-check: notes

  • Bucketing

    sorryhyun/anima_lora

    Free-fit native-shape bucketing — the token bands per edge tier, tier choice at preprocess time, the compiledynamicseq coupling and per-tier graph budget, and why training never needs --targetres.

    125 GitHub stars~932 tokensUpdated today
    Auto-check passed

  • Captions

    sorryhyun/anima_lora

    Caption pipeline — position-clause grammar (never hand-split a caption), make caption-autotag modes, make caption-position (v2 rewrite rules and gates), and the preprocess-stage wiring for both.

    125 GitHub stars~1.3k tokensUpdated today
    Auto-check passed

  • Custom Nodes

    sorryhyun/anima_lora

    The ComfyUI node map — which node lives in which standalone repo vs in-tree under customnodes/, where each is symlinked, and the vendor-sync rule for the vendor/ subsets.

    125 GitHub stars~470 tokensUpdated today
    Auto-check passed

  • Daemon

    sorryhyun/anima_lora

    Submit, monitor, and manage GPU jobs through the anima daemon (make daemon-, make gen, make run-status, MCP bridge, discovery).

    125 GitHub stars~2.1k tokensUpdated today
    Auto-check passed

Works with

Categories

Questions about Anime Tools

What does Anime Tools do?

The trainer ↔ animetools boundary — what the curation split moved out, the typed request/stage API the make targets build, the git-pin dev loop and its stale-venv trap, and the tests that guard the…. Anime Tools is an agent skill from sorryhyun/anima_lora. The trainer ↔ animetools boundary — what the curation split moved out, the typed request/stage API the make targets build, the git-pin dev loop and its stale-venv trap, and the tests that guard the contract.

When should I use Anime Tools?

Anime Tools fits situations like: development work in your project.

How do I install Anime Tools in Claude Code?

Run `npx skills add sorryhyun/anima_lora --skill anime-tools -a claude-code`. Or copy the skill folder (.claude/skills/anime-tools in sorryhyun/anima_lora) into .claude/skills/anime-tools in your project. Claude Code loads it when a task matches its description.

How do I install Anime Tools in Codex?

Run `npx skills add sorryhyun/anima_lora --skill anime-tools -a codex`. Or copy the skill folder (.claude/skills/anime-tools in sorryhyun/anima_lora) into .agents/skills/anime-tools in your project. Codex loads it when a task matches its description.

Can I use Anime Tools in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add sorryhyun/anima_lora --skill anime-tools -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/anime-tools, .gemini/skills/anime-tools, .github/skills/anime-tools and .opencode/skills/anime-tools in your project.

What does Anime Tools need to run?

Going by SKILL.md and its folder, Anime Tools needs the command-line tools its instructions call (uv, make and python). Our summary lists: Python 3.

Does Anime Tools access the network?

SKILL.md names 1 domain. As links in the text: github.com. This is read from the text; nothing was executed.

Is Anime Tools safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Anime Tools use?

Anime Tools is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Anime Tools use?

About 1.5k tokens (SKILL.md is roughly 5.9k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Anime Tools?

Skills that share tags, products or a category with Anime Tools: Finishing a Development Branch (obra/superpowers, 297k stars), Code Review Checklist (shareAI-lab/learn-claude-code, 78k stars), Code Design Rationale Investigator (cursor/plugins, 11k stars) and Contributor-First PR Merge (HKUDS/OpenHarness, 16k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Anime Tools?

sorryhyun (a GitHub user) maintains it in sorryhyun/anima_lora, which has 125 GitHub stars. The repository holds 12 skills in this directory. The repository was last updated on October 10, 2026.

Source: sorryhyun/anima_lora on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.