Skip to content
SkillModel-led2,960 tokens in use

dbs-report

Merges several saved diagnosis files for one project into a dated Markdown report you can send to a partner or keep for later review.

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

50tokens idle
2,960tokens in use
4,831tokens, 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

shelldetectedwrites filesdetected

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
  • Examples12/15
  • Output0/15
  • Size15/15
  • Resources7/15

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-report is a Chinese-language agent skill that works on the saved states left by the dbs-save skill. It reads every save file for a project, sorts them by time, and writes one new Markdown report in a fixed six-part layout.

The report lists how the user’s main complaint changed over time, the conclusions reached, the directions rejected, the open questions, the confirmed next step, and an index of the source files. Each line traces back to a field in a save file; nothing is summarized from memory.

What You Get

  • A Markdown report named with a timestamp and the project name, never overwriting an earlier one
  • A timeline of the user’s main complaint with the skill that produced each save
  • Conclusions merged and de-duplicated, newest first, with corrected ones marked
  • Rejected directions with the reasons, open questions, and the confirmed next step
  • An index table of every source file with date, title, status and path
  • A short receipt giving the file path, the number of saves merged and the date span

Common Use Cases

  • Sending a business partner a single document that sums up weeks of diagnosis sessions
  • Reviewing three weeks later how the problem framing changed
  • Reconciling the saved results with an outside advisor’s notes

What Makes It Unique

Every statement in the report comes from a field in a save file, and the only section where it may summarize is the evolution of the complaint, where it describes the path only. Conflicting conclusions are both kept, with the newer first and the older marked as corrected. Sensitive details the user saved are carried over unchanged, since masking belongs to the save step. It produces Markdown only.

Under The Hood

It is a prose-only skill with no scripts of its own. The agent reads the SKILL.md and a shared archive-contract reference. It reads the project’s save files from the archive folder (by default under the home directory, or the folder set in a small config file in the working directory), runs a shell command to create the reports folder if missing, and writes a new report file each run. The shared contract also names a short Python command for timestamps. It does not call web services, and it only reads the config, never changes it.

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