agentleFS
lgldlklgldlk/lgldlk-agent-skills/skills/weijincha-content-check/SKILL.md

weijincha-content-check

Skill3 starsChanged 5 months agoMITView on GitHub

Check Chinese inline text, local text files, and public HTTP(S) text-file links for prohibited, advertising, sexual, violent, political, platform-specific, and industry-specific risk terms through the Weijincha multi-scope endpoint. Use when Codex needs to review Xianyu listings, chat copy, attached text-like files, linked copy files, content for Xiaohongshu, Douyin, Kuaishou, Bilibili, WeChat, Taobao, JD, Pinduoduo, or e-commerce, health, beauty, education, and game compliance wording.

The file

We split this skill into the rules it gives an agent

10 rules. About 0 are common, 9 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈9

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use the bundled CLI to submit public-facing Chinese copy, local text files, or public text-file URLs to the selected risk scope and return …

    line 8
  2. Very specific:

    Run scripts/check_text.py with --text, --file, --url, or stdin.

    line 13
  3. Distinctive:

    When integrating into an application, call the provider from the backend, add a timeout, cache identical text, limit request frequency, and …

    line 16
  4. Distinctive:

    Read from stdin when the text is long:

    line 32
  5. Distinctive:

    Check a local text-like file:

    line 40
  6. Distinctive:

    Check a public text-file link:

    line 46
  7. Distinctive:

    Do not call the provider directly from browser code.

    line 65
  8. Distinctive:

    Do not depend on this undocumented endpoint as the sole publishing gate or represent it as an official Xianyu service.

    line 66
  9. Distinctive:

    Read references/api.md for endpoint details, supported scopes, response semantics, and known limitations.

    line 70
  10. Distinctive:

    Read references/files.md when reviewing local files or remote file links.

    line 71

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “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

Weijincha Content Check

Use the bundled CLI to submit public-facing Chinese copy, local text files, or public text-file URLs to the selected risk scope and return normalized JSON.

Workflow

  1. Select the narrowest applicable scope. Use xianyu for Xianyu titles, descriptions, and public chat templates.
  2. Run scripts/check_text.py with --text, --file, --url, or stdin.
  3. Inspect success, passed, words, and each scope result. Treat matches as review signals, not legal conclusions.
  4. Explain false-positive risk and manually review normal second-hand descriptions, brand names, and contextual quotations.
  5. When integrating into an application, call the provider from the backend, add a timeout, cache identical text, limit request frequency, and retain a local deterministic fallback.

Commands

Check one scope:

python scripts/check_text.py --scope xianyu --text '支持私下交易,拍前私聊确认。' --pretty

Combine several scopes:

python scripts/check_text.py --scope xianyu --scope ban --scope ad --text '待检测文案' --pretty

Read from stdin when the text is long:

python scripts/check_text.py --scope xianyu --pretty

Then provide the text on stdin.

Check a local text-like file:

python scripts/check_text.py --scope xianyu --file ./listing-copy.md --pretty

Check a public text-file link:

python scripts/check_text.py --scope xianyu --url 'https://example.com/listing.txt' --pretty

Repeat --file or --url to audit several sources. Long sources are split into bounded chunks and results include character and line locations.

Interpretation

  • success=false: The provider request or response failed. Do not interpret this as approval.
  • passed=true: The provider returned no matched words. Do not promise that the content will pass platform review.
  • passed=false: Review every matched phrase in context. Block only when local policy assigns a blocking severity.
  • Different calls can return different phrase grouping. Deduplicate matches and avoid exact-output snapshot assumptions.

Safety and Reliability

  • Submit only public listing copy. Never send card codes, passwords, cookies, tokens, personal data, or unredacted automatic-delivery payloads.
  • Inspect only text-like files. Reject binary files, private-network URLs, credential-bearing URLs, oversized sources, and unsupported encodings.
  • Do not call the provider directly from browser code; its CORS policy is restricted.
  • Do not depend on this undocumented endpoint as the sole publishing gate or represent it as an official Xianyu service.
  • Obtain provider permission before production use and keep the endpoint replaceable.
  • Validate scopes against the bundled allowlist; never interpolate arbitrary user input into the provider URL.

Read references/api.md for endpoint details, supported scopes, response semantics, and known limitations. Read references/files.md when reviewing local files or remote file links.

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.