Skip to content
SkillModel-led7,542 tokens in use

dbs-content-risk-check

Checks posts, scripts, captions and profiles sentence by sentence before publishing, telling machine-filter flags from real content problems and giving the smallest fix.

11,518 stars on the repository Custom / Unknowndontbesilent2025/dbskill · skills/dbs-content-risk-check
npx -y skills add dontbesilent2025/dbskill -g --all
View source

Install command as written in the repository's README

Skill label

measured from its files

What installing dbs-content-risk-check costs

216tokens idle
7,542tokens in use

Idle is what sits in context while the skill is installed. In use is its instructions, loaded when it runs. Counts are approximate, made offline with cl100k_base.

Model dependence Model-led

Script-ledMixedModel-led

Mostly instructions, so how well it works depends on the model that follows them.

This describes the skill's structure, how much of it is scripts. It does not say how well it works.

Touches

Nothing detected: no shell, network, file writes, git or credentials in its files.

Declared tags come from the skill's frontmatter; detected ones from its scripts, its code blocks and the commands it names in inline code. This is detected, not audited.

How it scores

Quality

How well the skill is written, counted from its own files: its frontmatter, its steps, examples, stated output and constraints, its size and whether every file it ships is one it names.How we score it →
Fair
  • Frontmatter8/15
  • Structure16/25
  • Examples9/15
  • Output0/15
  • Size11/15
  • Resources0/15

Security

Starts at 100 and loses points for each thing in the skill's files that a reader should look at first.How we score it →
Excellent

No findings.

Repository health

Pre-computed score based on development activity, maintenance, community, maturity, and trend momentum.How we score it →
Excellent
  • Development activity96
  • Maintenance100
  • Community64
  • Maturity28
  • Momentum40

dbs-content-risk-check is a pre-publish review for creators on Chinese social and content platforms. You submit a title, body, subtitles, voice-over, image text, a profile or a video description, and it points to the exact sentence or frame that may cause blocking, review or reduced reach.

It answers two separate questions: what an automatic filter may pick up, such as contact details, links, QR codes or listed words, and what is a real problem in the content, such as fraud, false claims or illegal lead generation. Strong opinions, emotion and personal style stay untouched. When asked about a video disclaimer, it also writes a three-line on-screen note.

What You Get

  • A two-line conclusion: what machine review may flag and whether the content has a real problem
  • A priority list of quoted passages with the concrete problem and the smallest edit
  • A list of passages that may trip machine review, with notes on likely false alarms
  • Facts to confirm before publishing, such as product category, qualification or sponsorship
  • An optional three-line on-screen note for a video, when the context calls for it
  • A statement of what was not checked

Common Use Cases

  • Checking a Xiaohongshu note for contact details, external links or ad structure before posting
  • Reviewing recruitment, order-taking or product posts for lead-generation and qualification issues
  • Reviewing medical, finance or education claims for unsupported promises
  • Asking what small caption to put on a video that states an opinion or personal experience

What Makes It Unique

Most keyword checkers return a list of flagged words. This Skill keeps machine-filter risk and content risk as two separate verdicts, demands a named mechanism before calling something a real violation, and keeps strong opinions and personal voice. It refuses to produce workarounds such as homophones or coded comments, never says a post can or cannot be published, and gives no legal opinion.

Under The Hood

This Skill is instructions only. The agent reads SKILL.md, which holds the review steps, the risk categories, the report format and the rules for on-screen notes, and applies them to what you submit. It has no scripts and writes no files. It does not open links in your draft unless you ask for the page to be checked, and it looks up a platform’s current official rules on the web only when the answer depends on them.

Join founders building with open source

Opinionated takes, migration guides, cost-saving tips, and insights from the open source ecosystem.

Subscribe on SubstackJoin 750+ subscribers