agentleFS
PapyGaragosPapyGaragos/skills/definition-of-done/SKILL.md

definition-of-done

Skill2 starsChanged 44 days agoMITView on GitHub

Say what counts as done for a task, goal or effort. Gives the few properties that decide success, each with the conditions that mark it reached, held, or good enough, without saying how to check them. Use when the user asks what should count as done, what success looks like, what the acceptance or exit criteria are, how they would know a goal is met, or wants a definition of done, and whenever another skill needs the success conditions of a goal.

10 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use tdd.

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.
  • GoogleSkill

    tdd

    Test-driven development. Use when the user wants to build features or fix bugs test-first, mentions "red-green-refactor", or wants integration tests.

    Also in 3 projects that use this

The file

We split this skill into the rules it gives an agent

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

  • 0

    Common

    Most of the 8 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Stop after the last entry, with no preamble or advice.

    line 38
  2. Distinctive:

    If the input names no outcome at all, so there is nothing whose success could be defined, say so in one line and stop.

    line 42

“Common” is measured: kept by at least 6 in 10 of the 8 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

Definition of done

A definition of done names the conditions that are true once a goal is met. It says what someone would find, not how they would look. These conditions are the end state, which people keep aiming at when the plan changes (ADP 6-0, 2026).

This skill decides what counts as done. It does not pick, write or run the checks that confirm it.

Steps

  1. Name the properties that decide success. Ask what would have to be true for the work to count, and keep only the properties where being wrong would matter. Most goals have one to three. Keep two properties apart when one could fail without the other, or when one is a matter of fact and the other a matter of judgment. Keep to the finished state: states that only matter on the way there belong to the plan, not to the definition of done.
  2. State each property as an effort. Call the Skill tool with "as-effort" and work out each property's default one-line form. The effort's slug becomes the property's name, and its end state is what the conditions describe. The effort lines stay out of the output, even though as-effort asks for them first. They shape the conditions, and the reader only needs the conditions. The kind sets their shape in step 3.
  3. Write the conditions in the shape of the kind.

    kindthe conditions
    achieveDone when: each condition that marks the state reached
    maintainHeld when: each condition that marks the state held
    optimizeMetric: what is measured, and the target

    Write each condition so that someone could find it true or false by looking: name what they would look at and what they would find. Words like works, clean or fast hide the condition instead of naming it. Leave out who or what does the looking, and how. When the input gives no target for a metric or threshold for a condition, infer the likeliest one and mark it (guessed). How many samples or runs a check takes belongs to the check, not here. When the property is a judgment of what is better or worth doing, such as taste, direction or priority, the condition is a decision: Done when: <who> decides or approves <what>.

What to produce

One entry per property:

<property>  Done when: <condition>; <condition>.
<property>  Held when: <condition>; <condition>.
<property>  Metric: <what is measured>, target <target>.

Stop after the last entry, with no preamble or advice. Another skill may read the output as it is, so keep this shape even when talking to a person.

When it doesn't fit

If the input names no outcome at all, so there is nothing whose success could be defined, say so in one line and stop.

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.