agentleFS
aya-talebaya-taleb/productivity-skills/skills/decision-log/SKILL.md

decision-log

Skill1 starsChanged 21 days agoMITView on GitHub

Logs a business decision in decisions.md with the context, the options you weighed, what you chose and why, what would change your mind, and a date to revisit. Also finds past decisions and lists the ones due for a second look. Use it when the user says "I decided to", "we're going with", "log this decision", "should I raise prices or not, help me write it down", "why did I pick Stripe again", "what decisions are due for review", "let's revisit the pricing call", or after a meeting where something big was agreed. Works for any business in founder-context.

46 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use product-brief, pre-mortem and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • erik-kroonSkill

    product-brief

    Turn conversation context, codebase findings, and product intent into a PRD-style brief without assuming a specific issue tracker. Use when the user wants a PRD, product brief, feature spec, or implementation-ready product artifact.

    Also in 3 projects that use this

  • Ingramradical235Skill

    pre-mortem

    Pre-mortem analysis, scenario planning, future state narratives, and contingency frameworks. Use before finalizing any Strategy Kernel, when planning major initiatives, or when preparing Plan B triggers.

    Also in 3 projects that use this

  • MicrosoftSkill

    retrospective

    Review how work went and improve the next round. USE WHEN asked for a retro, delays, or process lessons. DO NOT USE for code review or transcript replay.

    Also in 4 projects that use this

  • OpenAISkill

    meeting-prep

    Use when creating Public Equity Investing meeting prep briefs. Do not use for private diligence, IB, FP&A, legal, or scheduling-only tasks.

    Also in 3 projects that use this

  • Microsoft AzureSkill

    test

    Also in 3 projects that use this

  • HashiCorpSkill

    changelog

    Add a `.changelog/<PR_NUMBER>.txt` entry for a GitHub Pull Request, commit, and push.

    Also in 3 projects that use this

The file

We split this skill into the rules it gives an agent

7 rules. 0 are common, 4 are distinctive, and 3 are very specific to this project.

  • 0

    Common

    Most of the 39 edited versions we read keep these.

  • 4

    Distinctive

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Create the file if it is missing.

    line 18
  2. Distinctive:

    If the user is still deciding, help them lay out the options and the questions that decide it, but leave the choice to them.

    line 38
  3. Very specific:

    Never invent numbers, quotes or facts in the context.

    line 47
  4. Very specific:

    Add an Outcome (date): line under the entry: kept, changed, or reversed, and what was learned.

    line 57
  5. Distinctive:

    If changed, log the new decision as a new entry that points back to the old one.

    line 60
  6. Distinctive:

    Search decisions.md by business tag, keyword or date.

    line 64
  7. Distinctive:

    Keep the reply short and plain: sentences under 25 words, light formatting (a few bullets or one small table, little bold), no em or en …

    line 69

“Common” is measured: kept by at least 6 in 10 of the 39 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Decision log

Founders running several ventures make many calls a week and forget why. A short written record stops the same debate from happening three times, and a revisit date turns a gut call into something you can check.

File

decisions.md in the founder folder. python3 <this skill's folder>/../tasks/scripts/tasks.py context prints the folder; run it from the user's current folder (don't cd into the skill folder; the tool looks for ./founder/ where it runs). Newest entry at the top, under a # Decisions heading. Create the file if it is missing. Entry layout: references/entry-format.md.

Logging a decision

Get these from the user. If the user gave the decision and a reason, write the entry right away with {{placeholders}} for missing fields, and ask for those in the same reply. If they gave less, ask once, in one message, before writing.

  • Business: a slug from founder-context.
  • Decision: one sentence, what was chosen.
  • Context: why this came up now, in two or three lines.
  • Options: the choice plus what else the user says was on the table. Don't add options they didn't name. If they only had one option, write that down; it's useful later.
  • Why: the reasons in the user's words. Don't add reasons they didn't give.
  • What would change my mind: a signal that would make this the wrong call. This is what the revisit checks.
  • Revisit on: an absolute date. Suggest one that fits: about 2 weeks for a small test, 1 to 3 months for pricing or a hire, 6 months for a big direction. The user picks.
  • Reversible?: yes or no. A one-way door deserves a slower look.

If the user is still deciding, help them lay out the options and the questions that decide it, but leave the choice to them. Don't hint at a lean, suggest a default, or weigh one side's downsides more than the other's; asking you to decide is a sign they need the questions, not your vote. Don't write an entry until they choose. Stick to facts in the files or that the user gave; don't assume their state, permits or risks. Write the entry only once they choose. Where a decision has legal, tax or financial weight (a contract, an entity, an investment), say this is not legal, tax or financial advice and name who to ask (a lawyer, an accountant).

Never invent numbers, quotes or facts in the context. Missing numbers become placeholders like {{current monthly price}}.

Revisit

  • "What's due for review": list entries whose revisit date is today or earlier, oldest first. Say which ones are not due yet only if asked.
  • For each one, show the decision and "what would change my mind", then ask what happened. If the user already said what happened, compare it to that signal in one line ("41 a week, above the 35 line, so it held") and write the outcome.
  • Add an Outcome (date): line under the entry: kept, changed, or reversed, and what was learned. Never delete or rewrite the original reasons. The point of a log is seeing what you thought at the time.
  • If changed, log the new decision as a new entry that points back to the old one.

Finding past decisions

Search decisions.md by business tag, keyword or date. Quote the entry; don't paraphrase the reasons into something the user didn't say.

Reply style

Keep the reply short and plain: sentences under 25 words, light formatting (a few bullets or one small table, little bold), no em or en dashes (write times as 09:00-10:00). Don't guess at consequences ("the site may be down") that the files don't show. Finish with the one-line next step.

Handoff

End the reply with one line on what's next (for example "weekly-review will flag this on its revisit date.").

  • Reads: founder-context businesses, decisions.md.
  • Writes: decisions.md. Offer to add a "revisit <decision>" task to tasks.md on the revisit date (python3 <this skill's folder>/../tasks/scripts/tasks.py add ...).
  • Next: weekly-review reads revisit dates, so a due decision shows up in the week it's due.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.