agentleFS
Rome-1Rome-1/obsidian-claude-starter/.claude/skills/decision-record/SKILL.md

decision-record

Skill5 starsChanged 33 days agoMITView on GitHub

Records a decision with its context, the options considered, the reason for the choice, and what would make you revisit it, in a project's Decisions.md log or as a standalone decision note; and marks earlier decisions superseded instead of deleting them. Use when the user says "record this decision", "log that we decided X", "write an ADR", "why did we choose X?", "we're changing our mind on X", runs /decision-record, or when a meeting write-up or conversation produces a significant choice worth remembering.

9 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

5 rules. 0 are common, 3 are distinctive, and 2 are very specific to this project.

  • 0

    Common

    Most of the 6 edited versions we read keep these.

  • 3

    Distinctive

    This author's own choices. Worth a look.

  • 2

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    If the user only gave the decision, ask for the rest in one short message.

    line 21
  2. Distinctive:

    If this replaces an earlier decision:

    line 31
  3. Distinctive:

    Leave the old entry in place.

    line 33
  4. Very specific:

    Add a line to it: Superseded YYYY-MM-DD by [[...#New heading]]: <one-line reason>.

    line 34
  5. Very specific:

    Run python3 scripts/check_vault.py (it checks heading links like [[Decisions#2026-01-18 ...]]).

    line 41

“Common” is measured: kept by at least 6 in 10 of the 6 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 record

A decision record exists so that, months later, you (or an agent) can tell why something is the way it is, and whether the reason still holds.

1. Pin down the decision

Get, from the conversation or by asking:

  • The decision in one sentence, stated as what will be done.
  • Context: what forced the choice now.
  • Options considered, including "do nothing" when it was a real option, each with its main pro and con.
  • Why this one: the reason that tipped it. If the user cannot say, record that the reason is unclear rather than inventing one.
  • Revisit if: the evidence or change that would reopen it.

If the user only gave the decision, ask for the rest in one short message. Do not fill gaps with plausible-sounding reasons.

2. Choose where it goes

  • The project has a folder: add an entry at the top of Projects/<Name>/Decisions.md, under a heading ## YYYY-MM-DD <decision in a few words>.
  • The project is a single file, and the decision is small: a line in its ## Log.
  • The decision is big or cuts across projects: a standalone note from System/Templates/Decision.md, in the project folder or in Projects/.

3. Supersede, never delete

If this replaces an earlier decision:

  • Leave the old entry in place.
  • Add a line to it: **Superseded** YYYY-MM-DD by [[...#New heading]]: <one-line reason>.
  • For a standalone note, set status: superseded on the old one.

4. Link and check

  • Link the decision from the project's log, and from the meeting note it came from, if any.
  • Set updated: on every note you change.
  • Run python3 scripts/check_vault.py (it checks heading links like [[Decisions#2026-01-18 ...]]).

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.