agentleFS
Grit-77Grit-77/done-is-a-claim/skills/acceptance-design/SKILL.md

acceptance-design

Skill33 starsChanged 35 days agoApache-2.0View on GitHub

Define executable acceptance for a bounded task before work or dispatch. Use when writing a worker brief, selecting success checks, or repairing an acceptance contract that cannot test the intended outcome.

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

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈5

    Distinctive · estimate

    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:

    Test that it launches on the target environment before dispatch.

    line 21
  2. Distinctive:

    Do not depend on an undocumented interpreter or hidden login state.

    line 23
  3. Distinctive:

    If a worker cannot run a required live-service or platform check, assign it to a suitable verifier and retain it as required.

    line 38
  4. Distinctive:

    Put the exact commands, inputs, preservation requirements and stop conditions in the brief.

    line 44
  5. Distinctive:

    If the check is wrong or unrunnable, report the mismatch before continuing dependent work.

    line 46

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “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

Acceptance design

Acceptance connects the user's outcome to an observable check. Design it before implementation, and give the worker the exact contract the reviewer will use. These are operational instructions, not measured failure-rate claims.

Define the experiment

  1. Name the outcome, affected boundary and scope. Check the command, loader, artifact or user flow actually affected; a helper test alone may miss the boundary defect. Identify what the check deliberately leaves untested.
  2. Name the target host or platform, runner, dependencies, permissions and inputs. Confirm required source files exist in the delivered base and travel with the brief. Distinguish existing input paths from files to create.
  3. Record the exact command, expected observable result and evidence location. Test that it launches on the target environment before dispatch. Prefer a repository script when shell quoting or setup would make the command fragile. Do not depend on an undocumented interpreter or hidden login state.
  4. Scope every planned write, including tests and generated artifacts. Make predecessors available in the delivered base, or route the task to a checkout that contains them. A named branch is not necessarily accessible.

Separate change from preservation

  • For a defect or new behavior, demonstrate that the check rejects the old behavior for the intended reason when feasible. A missing tool or test path is not that demonstration. If a baseline run is unavailable, state the gap.
  • For preservation, establish the existing behavior and require it to remain valid. Do not demand a failing baseline for behavior meant to stay unchanged.
  • Vary meaningful inputs and inspect outputs where the contract depends on them. A mocked success, repeated implementation constant or file's existence may prove less than the required outcome. Include relevant regression scope.
  • If a worker cannot run a required live-service or platform check, assign it to a suitable verifier and retain it as required. Local success cannot stand in for the missing boundary check.

Freeze and correct the contract

Put the exact commands, inputs, preservation requirements and stop conditions in the brief. Worker and reviewer use the same contract on identified trees. If the check is wrong or unrunnable, report the mismatch before continuing dependent work. The task owner records the old and replacement contracts, the reason and how the replacement still tests the intended outcome. Revalidate the replacement before resuming; do not silently lower a threshold, remove a failing case or relabel a failure as success. A changed user requirement is a scope change to record, not evidence that the old contract passed.

Acceptance is designed when it can distinguish the required outcome from a plausible failure, runs where assigned, and preserves the agreed boundaries.

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.