brief-to-work-order
Skill0 starsChanged 48 days agoMITView on GitHub
Turn a request, rough brief, or agreed plan into a bounded work order with clear scope, deliverables, ownership, and checkable acceptance criteria. Use when someone is ready to assign execution. Do not use while the direction is still undecided (use Decision Deliberation) or to record finished or paused work (use Nutshell or Handoff).
In this repository10 other files it gives its agents.
- decision-deliberationSkill
- eval-viability-checkSkill
- evidence-checkSkill
- grounded-researchSkill
- handoffSkill
- nutshellSkill
- red-team-reviewSkill
- retrievable-writingSkill
- skill-or-notSkill
- source-briefSkill
The file
We split this skill into the rules it gives an agent
6 rules. About 0 are common, 6 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈6
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
- Distinctive: line 41
Do not treat a work order as permission for external actions that the user has not authorized.
- Distinctive: line 48
Read the work order as the implementer.
- Distinctive: line 53
Do not convert a suggestion into an approved decision.
- Distinctive: line 54
Do not hide uncertainty inside confident scope language.
- Distinctive: line 55
Do not write acceptance criteria such as "works well" or "tests pass" unless the observable condition and relevant test are named.
- Distinctive: line 57
Do not broaden the work just to make the document feel complete.
≈ “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.
Files
Brief To Work Order
Purpose
Convert intent into an assignment someone can execute and a requester can judge. Preserve decisions already made. Surface missing choices that could change the work instead of silently making them.
When to use
- A request, brief, or agreed plan is ready to be handed to someone for execution.
- A task keeps being misunderstood because its scope or finish line is vague.
- Work is being passed between people or agents and needs checkable acceptance criteria.
When not to use
- The direction is still undecided. Use Decision Deliberation first; a work order should not make the strategic choice.
- The task is small enough that one clear sentence already covers it.
- Someone wants a status update or a record of finished work. Use Handoff or Nutshell instead.
Workflow
- Identify the intended outcome, audience or user, current state, and reason for the work.
- Separate agreed decisions from assumptions and unresolved choices. Ask a focused question when an unresolved choice would materially change scope, cost, risk, or deliverables. Otherwise state a reasonable assumption.
- Define the smallest useful scope. Say what is included and explicitly name important exclusions.
- Name the owner, inputs, dependencies, constraints, and required authority. Do not treat a work order as permission for external actions that the user has not authorized.
- Describe concrete deliverables and atomic acceptance criteria. Each criterion should be one observable pass/fail check. Split compound criteria. A proposed check is not a completed result.
- Include verification and handoff expectations that fit the risk. Do not add process or ceremony without a reason.
- Read the work order as the implementer. Remove ambiguity about what to do next, what not to do, and how completion will be recognized.
Guardrails
- Do not convert a suggestion into an approved decision.
- Do not hide uncertainty inside confident scope language.
- Do not write acceptance criteria such as "works well" or "tests pass" unless the observable condition and relevant test are named.
- Do not broaden the work just to make the document feel complete.
Output shape
# [Outcome-oriented title]
## Goal
[Who needs what outcome and why]
## Current state and decisions
[Relevant facts and settled choices]
## Scope
### Included
- [Bounded work]
### Out of scope
- [Explicit exclusions]
## Deliverables
- [Concrete artifact or result]
## Owner and dependencies
- Owner:
- Inputs:
- Dependencies:
- Constraints or required authority:
## Acceptance criteria
- [One binary, observable check per line]
## Verification and handoff
[Checks, evidence, and recipient needs]
## Open questions
[Only decisions that still need an answer]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
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.

