Skip to content
SkillModel-led4,215 tokens in use

dbs-decision

Builds a local folder of Markdown files for a long-running decision area, keeping facts, patterns, snapshots and open questions apart so later sessions pick up where you stopped.

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

78tokens idle
4,215tokens 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
  • Examples12/15
  • 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-decision sets up a local knowledge project for any area you want to follow over months: a business, a relationship, health, a career, learning or investing. It does not decide for you. It records what happened, what you can infer, how things stood at a point in time and what remains unanswered, each in its own folder.

Each time you return with news, it reads the current-state file, gives an overall read with two or three observations and next directions, and updates the right files. Big forks such as a pricing change are filed as decision events, and outcomes are added to them later. The Skill text is in Chinese and replies in your language.

What You Get

  • A project folder with four layers: facts, patterns, snapshots and open questions, plus usage notes and a current-state file
  • Source tags that separate your words, the AI’s guesses and conclusions, and later corrections
  • Decision event files for big forks, with a result section filled in when the outcome is known
  • Dated snapshots of how things stood, which are never edited afterward
  • An optional privacy mode that stores real names only in a private mapping file

Common Use Cases

  • Following a business decision about pricing, partners or product structure over several months
  • Keeping a running record of a health or career question and revisiting it
  • Recording results of past decisions so patterns show up over time
  • Resuming a long discussion in a new chat by reading the saved files

What Makes It Unique

Most decision tools keep a list of choices. This one keeps four layers with different edit rules: facts are append-only, patterns get dated corrections, snapshots are frozen and open questions are cleared once answered. It requires every AI inference to carry a tag, keeps hindsight from rewriting earlier judgment, and stops giving options when you loop between asking and rejecting.

Under The Hood

This Skill is instructions only, with no scripts. The agent creates and edits local Markdown files, by default under a .dbs/decisions folder in your home directory, or in a decisions folder inside the current project if you ask for that. It reads the current-state file at the start of each session and updates it at the end. It runs no commands and makes no web requests. Notes may hold sensitive details, so privacy mode and keeping the folder out of commits are part of its guidance.

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