Skip to content
SkillModel-led4,438 tokens in use

dbs-benchmark

Finds a competitor worth copying with a five-filter test, maps how it gets and keeps customers, and sets a copying path that cuts out personal excuses.

11,518 stars on the repository Custom / Unknowndontbesilent2025/dbskill · skills/dbs-benchmark
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-benchmark costs

64tokens idle
4,438tokens in use
8,488tokens, up to

Idle is what sits in context while the skill is installed. In use is its instructions, loaded when it runs. Up to adds the files its instructions link. 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 →
Good
  • Frontmatter8/15
  • Structure22/25
  • Examples9/15
  • Output0/15
  • Size15/15
  • Resources11/15

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-benchmark helps you choose one business to imitate and work out how it makes money. It screens candidates on whether they earn well, whether you can understand them and whether you can copy them, then lays out the chain from getting customers to repeat sales and a step-by-step path to match it.

It also checks your own reasoning. When you say a model does not suit you or has nothing to do with your interests, the Skill points that out as noise and asks which step you cannot do. The Skill text is in Chinese and replies in the language you use.

What You Get

  • A first-choice benchmark with a one-sentence reason
  • A screening table scoring each candidate on profit, understandability and copyability
  • The chain of the chosen benchmark: customer acquisition, conversion, delivery and repeat purchase
  • A numbered path to imitate it, plus a list of facts still to verify
  • A table of the noise in your own statements, quoted back to you
  • A detail-by-detail comparison of you against the benchmark when you already picked one

Common Use Cases

  • Choosing which competitor to copy before launching a course or a store
  • Checking whether you really match a benchmark’s price, packaging, content format and sales script
  • Running a content or ad account on Xiaohongshu, Douyin or Bilibili and aligning with proven posts
  • Replying to the thought that a benchmark does not fit your personality

What Makes It Unique

Most competitor-analysis skills compare features or market size. This one rests on a stated rule: in the zero-to-one stage, copy first and differentiate later. It judges candidates only on profit, marks any revenue estimate as a guess with the basis written out, and treats each difference from the benchmark as something you must justify. Differences without a reason are changed back to match.

Under The Hood

The agent follows the instructions in SKILL.md and reads reference files for output templates, cases and platform tips when needed. It has no scripts of its own. When the host has a search tool, the Skill tells the agent to look up public information about candidates and cite sources; without one it gives a method and asks you for links. It writes no files.

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