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

Subject Audit

by yjl9903 in yjl9903/AnimeGarden

Review AnimeGarden animation resource-to-Bangumi Subject bindings for a specified time range, identify missing or incorrect bindings, and report evidence-backed Subject search improvements.

AGPL-3.0Auto-check: notesBackend & APIs

Install Subject Audit

skills CLI
$ npx skills add yjl9903/AnimeGarden --skill subject-audit -a claude-code

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

GitHub CLI
$ gh skill install yjl9903/AnimeGarden subject-audit --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/yjl9903/AnimeGarden.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.agents/skills/subject-audit .claude/skills/subject-audit && 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
subject-audit
GitHub stars
1.3k
Token cost
~1.6k tokens
SKILL.md length
720 words
Files
7 (incl. scripts, references)
Skills in repo
3
Repo updated
First seen
Licence
AGPL-3.0

At a glance

Review AnimeGarden animation resource-to-Bangumi Subject bindings for a specified time range, identify missing or incorrect bindings, and report evidence-backed Subject search improvements.

  • Works in 6 steps: Collect the audit evidence → Attribute every resource → Check the target Subject's effective match → …
  • Manager-side audits
  • Runs JavaScript scripts from its folder; calls node; needs ADMIN_SECRET
  • Online binding corrections are optional and require explicit user confirmation

What it does

Subject Audit is an agent skill from yjl9903/AnimeGarden. Review AnimeGarden animation resource-to-Bangumi Subject bindings for a specified time range, identify missing or incorrect bindings, and report evidence-backed Subject search improvements. Use for manager-side audits; online binding corrections are optional and require explicit user confirmation.

Its SKILL.md is about 1.6k tokens, which your agent loads only when the skill is triggered. The skill folder holds 8 other files, including scripts and reference files (for example `references/audit-rules.md`).

It sits in Backend & APIs. It works with OpenAPI. The repository describes itself as: 動漫花園 镜像站 | 动画 BT 资源聚合站 | 动画 BT 资源开放接口. The licence is AGPL-3.0.

When your agent uses it

  • Manager-side audits
  • Online binding corrections are optional and require explicit user confirmation

Example prompts

  • “/subject-audit”

Requirements

  • Node.js
  • A credential in ADMIN_SECRET

Workflow steps

6 steps, taken from the step headings in SKILL.md.

  1. Collect the audit evidence
  2. Attribute every resource
  3. Check the target Subject's effective match
  4. Diagnose Subject search coverage
  5. Optionally apply confirmed resource bindings
  6. Report the findings

What it can do on your machine

Read from SKILL.md and the folder at commit c1bd61c. 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

    Ships 5 files in scripts/ (JavaScript), which the agent can run.

    Shell commands in SKILL.md call:

    • node

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

  • Network

    No URLs in SKILL.md.

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

  • Credentials

    Names these keys or tokens, usually read from environment variables:

    • ADMIN_SECRET

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

Context cost

Subject Audit loads about 1.6k tokens when it runs, and up to ~5.5k if it reads all its reference files. Until then it costs about 78 tokens; SKILL.md has 720 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~78
When it runs · the whole SKILL.md, loaded when a task matches
~1.6k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~5.5k

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: notes

The automated check noted patterns worth knowing about, such as sudo or a known installer.

  • NoteMentions a .env fileSKILL.md:107
    ds `ADMIN_SECRET` from the project root `.env` and records server results in a journal.

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); the scripts in this folder are not scanned.

SKILL.md

The full file from yjl9903/AnimeGarden at commit c1bd61c, republished under its AGPL-3.0 licence (© yjl9903). 720 words, ~1,585 tokens.

Download SKILL.mdSave it as .claude/skills/subject-audit/SKILL.md (or your agent's skills folder). This skill also uses 6 other files; get the full folder from GitHub.
name
subject-audit
description
Review AnimeGarden animation resource-to-Bangumi Subject bindings for a specified time range, identify missing or incorrect bindings, and report evidence-backed Subject search improvements. Use for manager-side audits; online binding corrections are optional and require explicit user confirmation.

AnimeGarden Subject Audit

Review every animation resource in a requested time range, determine which registered Bangumi Subject it most likely belongs to, and assess both its current binding and the target Subject's search coverage. The current subjectId and search configuration are evidence about system state rather than the basis for attribution.

The standard path produces a read-only audit report. An optional confirmed step can apply reliable target Subject IDs to current online resources before the report is finalized.

Read references/audit-rules.md for the attribution standard, search diagnosis, and report format.

Workflow

1. Collect the audit evidence

Create a fresh task-scoped temporary directory and run from this skill directory:

bash
node scripts/fetch-resources.mjs --start <ISO timestamp> --end <ISO timestamp> --output <tmp-dir>/resources.jsonl
node scripts/fetch-subjects.mjs --output <tmp-dir>/subjects.jsonl

The first file contains all animation resources in the interval and their current subjectId; the second contains the complete registered Subject index. Confirm the declared counts, then review the resource file in manageable batches until every record has been assessed.

2. Attribute every resource

Identify the work, installment, and episode from the complete release context. For an existing binding, evaluate the bound Subject alongside other plausible registered Subjects. For an unbound resource, search the registry using the title forms suggested by the release.

Weigh the evidence that is informative for that resource: names and aliases, season or part context, episode progression, publication timing, related Subjects, and neighboring releases. Their relevance varies by title family, so attribution follows the explanation that best fits the whole release context. Record uncertainty when the evidence does not support one clear Subject.

When several registered installments share a practical title or continue episode numbering, fetch the direct predecessor and sequel relations for the plausible candidates:

bash
node scripts/fetch-subject-relations.mjs --subject-ids <id[,id...]> --output <tmp-dir>/subject-relations.jsonl

Use this optional evidence to navigate quarter and part boundaries in long-running families such as ONE PIECE and BLEACH. Relations narrow the candidate set; the resource title, episode sequence, and publication context still determine the attribution.

3. Check the target Subject's effective match

For every resource with a reliably established target, write one record per line to <tmp-dir>/targets.jsonl:

json
{ "provider": "mikan", "providerId": "123", "targetSubjectId": 456 }

When the target list is non-empty, run the production-equivalent match check:

bash
node scripts/check-subject-matches.mjs --resources <tmp-dir>/resources.jsonl --subjects <tmp-dir>/subjects.jsonl --targets <tmp-dir>/targets.jsonl --output <tmp-dir>/matches.jsonl

The result answers whether the selected target Subject matches each resource under the current production title normalization and search semantics. It also identifies the exact failed condition: an include miss or formatting-only near miss, missing keywords, matched excludes, and violated time boundaries. Combine those diagnostics with the current subjectId to describe the present search coverage and binding state. This check measures the current search configuration; the release evidence from step 2 remains the basis for attribution.

Show full SKILL.md (308 more words)Show less
4. Diagnose Subject search coverage

Inspect both coverage and selectivity in the target Subject's current search settings and, for a wrong existing binding, the currently bound Subject as well. A successful match confirms coverage but can still be too broad; short or common terms need supporting context from actual release families. Recommend the focused include, keywords, exclude, after, or before changes that best separate the current Subjects. Time signals may complement titles when resource publication ranges provide a reliable present-day boundary.

Treat include-near-miss as an inspection hint: the include becomes a substring after ignoring spaces, punctuation, and symbol separators. The production result remains unchanged, and the safe search adjustment depends on the complete release family.

5. Optionally apply confirmed resource bindings

If the user asks to apply reliable target bindings before the Subject search configuration is updated, generate a preview:

bash
node scripts/patch-resource-bindings.mjs --input <tmp-dir>/matches.jsonl --output <tmp-dir>/binding-patch-preview.jsonl

Present the exact resource count and current-to-target Subject mappings from the preview. After the user explicitly confirms that patch set, apply that preview file with the confirmed count:

bash
node scripts/patch-resource-bindings.mjs --input <tmp-dir>/binding-patch-preview.jsonl --output <tmp-dir>/binding-patch-results.jsonl --apply --expected-count <confirmed count>

Apply mode reads ADMIN_SECRET from the project root .env and records server results in a journal. Keep the audit read-only until the user confirms the exact previewed patch set.

6. Report the findings

Write the report in Chinese using the tables defined in the reference. Include the scan interval, summary counts, linked resource examples, and the correct numeric Subject ID with its Bangumi link whenever attribution is reliable. Show both current and correct Subjects for a wrong binding, and state the remaining ambiguity for unresolved cases.

When the optional online patch ran, include its changed, unchanged, and failed counts and identify any resource that was not completed.

Present Subject search recommendations as an unranked list and keep them focused on current title patterns and Subject search fields.

Compare with an earlier round when an actual prior evidence snapshot is available.

© yjl9903, AGPL-3.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 6 other files (scripts, references) in .agents/skills/subject-audit of yjl9903/AnimeGarden.

  • SKILL.md
  • references/audit-rules.md
  • scripts/check-subject-matches.mjs
  • scripts/fetch-resources.mjs
  • scripts/fetch-subject-relations.mjs
  • scripts/fetch-subjects.mjs
  • scripts/patch-resource-bindings.mjs

Open the folder on GitHubat commit c1bd61c

Compare with similar skills

Subject Audit 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.

Subject Audit compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Subject Audit this skillyjl9903/AnimeGarden1.3k—~1.6kAutomated safety check: NotesAGPL-3.0
ToolJet Marketplace Plugin BuilderToolJet/ToolJet41k—~2.1kAutomated safety check: PassAGPL-3.0
Step Partsearthtojake/text-to-cad19k1 repos~1.5kAutomated safety check: PassMIT
OpenAPI to MCP Servermcp-use/mcp-use11k—~5.2kAutomated safety check: PassApache-2.0
Use Yaakmountain-loop/yaak19k—~1.9kAutomated safety check: PassMIT
API DesignerJeffallan/claude-skills12k1 repos~2kAutomated safety check: PassMIT

Similar skills

  • ToolJet Marketplace Plugin Builder

    ToolJet/ToolJet

    Turns an API description, such as an OpenAPI file or a Postman collection, into a connector plugin for ToolJet's marketplace and checks it with the repo's validator.

    41k GitHub stars~2.1k tokensUpdated yesterday
    Backend & APIsAuto-check passed

  • Step Parts

    earthtojake/text-to-cad

    Find, evaluate, and download common purchasable CAD parts from step.parts, including named off-the-shelf actuators, servos, motors, electronics boards, connectors, screws, bolts, nuts, washers…

    19k GitHub starsUsed in 1 repo~1.5k tokens
    Backend & APIsAuto-check passed

  • OpenAPI to MCP Server

    mcp-use/mcp-use

    Turns an OpenAPI or Swagger spec into an MCP server with the mcp-use TypeScript SDK, mapping each operation to a tool, wiring auth, testing and deploying.

    11k GitHub stars~5.2k tokensUpdated 2 days ago
    Backend & APIsAuto-check passed

  • Use Yaak

    mountain-loop/yaak

    A skill your agent uses when the user mentions Yaak, a Yaak workspace, or the yaak command, or asks to call, hit, or smoke test HTTP/REST endpoints, save or organize API requests for reuse or manual…

    19k GitHub stars~1.9k tokensUpdated today
    Backend & APIsAuto-check passed

  • API Designer

    Jeffallan/claude-skills

    Designs REST and GraphQL APIs from resource modeling to an OpenAPI 3.1 contract, with versioning, pagination and RFC 7807 error handling.

    12k GitHub starsUsed in 1 repo~2k tokens
    Backend & APIsAuto-check passed

  • Pangolin CRUD Endpoints

    fosrl/pangolin

    Use whenever asked to add, create, or scaffold a CRUD endpoint, router, or entity in this repo's server (create/list/get/update/delete handlers, new…

    23k GitHub stars~461 tokensUpdated yesterday
    Backend & APIsAuto-check passed

More from yjl9903/AnimeGarden

  • Animegarden

    yjl9903/AnimeGarden

    Anime Garden is an anime torrent resources aggregation platform for 動漫花園, 蜜柑计划, 萌番组, and ANi.

    1.3k GitHub stars~824 tokensUpdated yesterday
    Auto-check passed

  • Yuc

    yjl9903/AnimeGarden

    Scrape quarterly anime lineups from https://yuc.wiki/ and output structured results for a specified or latest quarter.

    1.3k GitHub stars~613 tokensUpdated yesterday
    Auto-check passed

Works with

Categories

Questions about Subject Audit

What does Subject Audit do?

Review AnimeGarden animation resource-to-Bangumi Subject bindings for a specified time range, identify missing or incorrect bindings, and report evidence-backed Subject search improvements. Subject Audit is an agent skill from yjl9903/AnimeGarden. Review AnimeGarden animation resource-to-Bangumi Subject bindings for a specified time range, identify missing or incorrect bindings, and report evidence-backed Subject search improvements.

When should I use Subject Audit?

Subject Audit fits situations like: manager-side audits; online binding corrections are optional and require explicit user confirmation.

How do I install Subject Audit in Claude Code?

Run `npx skills add yjl9903/AnimeGarden --skill subject-audit -a claude-code`. Or copy the skill folder (.agents/skills/subject-audit in yjl9903/AnimeGarden) into .claude/skills/subject-audit in your project. Claude Code loads it when a task matches its description.

How do I install Subject Audit in Codex?

Run `npx skills add yjl9903/AnimeGarden --skill subject-audit -a codex`. Or copy the skill folder (.agents/skills/subject-audit in yjl9903/AnimeGarden) into .agents/skills/subject-audit in your project. Codex loads it when a task matches its description.

Can I use Subject Audit 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 yjl9903/AnimeGarden --skill subject-audit -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/subject-audit, .gemini/skills/subject-audit, .github/skills/subject-audit and .opencode/skills/subject-audit in your project.

What does Subject Audit need to run?

Going by SKILL.md and its folder, Subject Audit needs JavaScript for the scripts in its folder, the command-line tools its instructions call (node) and credentials named ADMIN_SECRET. Our summary lists: Node.js; A credential in ADMIN_SECRET.

Does Subject Audit access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Subject Audit safe to install?

Our automated static check of SKILL.md found notes only (mentions a .env file), nothing it rates as a warning. It is not a guarantee. The check reads SKILL.md only: the scripts in the folder are not scanned, so read them before running anything.

What licence does Subject Audit use?

Subject Audit is published under the AGPL-3.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Subject Audit use?

About 1.6k tokens (SKILL.md is roughly 6.3k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 3.9k tokens, read only when the agent opens those files.

What are the alternatives to Subject Audit?

Skills that share tags, products or a category with Subject Audit: ToolJet Marketplace Plugin Builder (ToolJet/ToolJet, 41k stars), Step Parts (earthtojake/text-to-cad, 19k stars), OpenAPI to MCP Server (mcp-use/mcp-use, 11k stars) and Use Yaak (mountain-loop/yaak, 19k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Subject Audit?

yjl9903 (a GitHub user) maintains it in yjl9903/AnimeGarden, which has 1,331 GitHub stars. The repository holds 3 skills in this directory. The repository was last updated on October 10, 2026.

Source: yjl9903/AnimeGarden on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.