Skip to content
SkillModel-led3,541 tokens in use

dbs-goal

Turns a vague wish into a goal an agent can act on and check, asking at most one question at a time and only when the answer changes the work.

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

95tokens idle
3,541tokens 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
  • Structure22/25
  • Examples9/15
  • Output0/15
  • Size15/15
  • 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-goal is a Chinese-language agent skill that audits a goal before work starts. It reads everything already said in the conversation, pulls out the deliverable, audience, scope, specs and finish line, and asks only about gaps that would change the result.

When the goal is already clear it returns a short goal card and stops. When it is only a wish, such as “build a personal brand”, it asks one narrowing question per turn until the goal can be written down and checked. It never invents numbers, deadlines or success criteria for you.

What You Get

  • A short goal card with deliverable, audience, scope, completion condition, assumptions and next step
  • One specific question when a single gap would change the deliverable, route or acceptance
  • A fuller audit with the user’s own words quoted, vague terms marked, and a checklist of completion conditions
  • Assumptions labelled as assumptions, never presented as confirmed facts

Common Use Cases

  • Starting a report task where audience, data scope, format and length are already given, and getting a go-ahead with no extra questions
  • Narrowing ‘I want a personal brand’ to an observable change such as paid consultations
  • Checking a vague phrase like ‘a deeper analysis’ to see which part changes what gets delivered

What Makes It Unique

It treats questions as costly. Before asking, it checks whether the answer already exists in context, can be inferred safely, or would change the deliverable; if not, it does not ask. Words are tested by whether they change the deliverable, route or acceptance result, which is stricter than asking if a sentence still reads well without them. The user keeps ownership of numbers, deadlines and success standards, and the skill says what it assumed.

Under The Hood

It is a prose-only skill with no scripts and no reference files; the agent reads the SKILL.md and a small agent metadata file. It works from the conversation and any attachments you provide. It does not run commands, call web services or write files, and it does not carry out the goal it clarifies.

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