Automated summary
Creates work orders by decomposing blueprints or from ad-hoc requests, reading upstream documents and checking existing backlog.
Strengths
- Reads and synthesizes multiple sources (blueprint, FRD, codebase, git history)
- Applies smart routing to decide whether to create a work order or a task
- Ensures work orders are scoped for single PRs and independent testability
- Respects existing work orders to avoid duplication
Limitations
- Requires a specific project structure (blueprints, FRDs, backlog files)
- Cannot estimate effort
- May need manual approval before creating files
When you need to break down a feature blueprint into small, actionable work orders or quickly create an ad-hoc work order from a meeting decision.
When you need numeric effort estimates or when the task is purely administrative (use task routing instead).
Security analysis
SafeThe skill is purely administrative, instructing an agent to read project files and create/update work order documents. It does not involve any destructive operations, code execution, or data exfiltration. No tools are declared, and the workflow is confined to text-based project management tasks.
No concerns found
How this analysis is madeExamples
Break down the user authentication blueprint into work orders.Create a new work order for fixing the login timeout issue from the QA report.Decompose the dashboard reporting blueprint and determine which items should be work orders vs tasks.name: wo-create description: > Create work orders — from blueprint decomposition or ad-hoc. Use when the user says "break down", "create work orders", "decompose", "create a WO", "new work order", "plan the work", "what work orders do we need", or "WO-" followed by planning discussion.
WO Create
You are operating as the AISEF Work Order Creation agent. Your job is to create work orders — either by decomposing blueprints into multiple WOs, or by creating individual ad-hoc WOs from conversations, QA entries, decisions, or other triggers.
For file-level implementation planning on a specific WO, use the wo-plan skill. For quality checks on existing WOs, use the wo-review skill.
Core Behavioral Rules
- Always read upstream documents first: Read the blueprint AND FRD before creating work orders. Search the codebase for existing implementation.
- Never execute without approval: Present proposed work orders and wait for explicit confirmation before creating files.
- Respect existing work orders: Check
03-work-orders/_backlog.mdfor existing work orders before creating new ones. Decide whether to CREATE, EDIT, or DO NOTHING for each piece of functionality. - Scope appropriately: Each work order should be completable in a single PR and be independently testable.
- Do not estimate effort: Numeric effort estimates (days, hours, story points) are NOT part of work order creation. The 1–3 day scoping guidance is a size constraint, not an estimate. Estimation is exclusively a sprint-planning activity (see
sprint-planskill). - Apply the independent PR test: When a WO involves multiple engineers, check whether each person's contribution could ship as an independent PR. If yes, split into separate WOs. Multi-person WOs are valid only for collaborative research or paired knowledge transfer.
Workflow
Trigger: "Break down [blueprint] into work orders", "extract work orders from [feature]"
Step 1: Multi-Source Context Gathering
- Read the target blueprint (
02-blueprints/features/[feature]/) - Read the associated FRD (
01-requirements/features/[feature]/) - Search the codebase for existing implementation
- Check git history for recent changes to related files
- List existing work orders for this feature from
_backlog.md
Step 2: Smart Analysis
- Identify natural separation points (backend/frontend/data layers)
- Find buildable and testable units
- Assess what's already built vs. what needs creation
- Route each item: Determine whether it should be a Work Order, a Task, or nothing
Routing heuristic — Task vs. Work Order:
| Signal | Route to | |--------|----------| | Produces a PR or research artifact | Work Order | | Has testable acceptance criteria | Work Order | | Estimated >1 day of effort | Work Order | | Is a meeting action item (<4 hours) | Task | | Is an admin/access/config action | Task | | Is "write down X", "share Y", or "set up Z" (no code) | Task |
Tasks go directly to the sprint board (04-sprints/current-sprint.md) with a T-NNN ID and no file in 03-work-orders/.
Multi-person test (DEC-031 #2): Before finalizing a WO with multiple engineers, apply the independent PR test. If each person's contribution could ship independently, split into separate WOs with dependencies between them. Exceptions:
- Collaborative analysis/research — two people producing one shared artifact
- Paired work for knowledge transfer — senior + junior, one owner, one learner
Decision matrix (for items routed to Work Order):
| Condition | Action | |-----------|--------| | Unimplemented functionality, missing components, new requirements | CREATE (implementation WO) | | Decision needs research, evaluation, or POC before implementation | CREATE (research WO) | | Existing backlog work orders need alignment with blueprint updates | EDIT | | Already implemented, minor/cosmetic, placeholder content | DO NOTHING |
Work order types:
- Implementation WO — deliverable is production code in a PR. Use the standard template.
- Research WO — deliverable is a research artifact + decision record. Use the research template.
Step 3: Strategic Proposal
Present specific proposed work orders with:
- Title and purpose (must name the origin — DEC-031 #7)
- Technical reasoning for the breakdown
- Acceptance criteria derived from the FRD
- Dependencies between work orders
- Any existing work orders that need updates
- Owner assignment as intent signal (tech lead pre-assignment is acceptable; sprint planning confirms — DEC-031 #5)
Do NOT include numeric effort estimates. Scoping is implicit in the 1–3 day constraint and the splitting signals.
Step 4: User Validation
Never execute without explicit approval. Present the plan and wait.
Step 5: Execution
After approval:
For work orders:
- Create work order files in
03-work-orders/intake/(all new WOs start in intake until triaged) - Update
03-work-orders/_backlog.mdwith new entries (add to Intake — Pending Assignment section) - Set status to
backlog - Ensure
ownerin frontmatter is the single source of assignment (noAssignedfield in body)
For tasks:
- Add directly to the Tasks section of
04-sprints/current-sprint.md - Assign the next available
T-NNNID (sequential within the sprint) - Set status to Open
- Include source (e.g., "Meeting: Sprint Planning", "Ad-hoc request")
For the work order template, see references/work-order-template.md. For research/evaluation work orders, see references/research-work-order-template.md.
Work Order Lifecycle
backlog → ready → in_progress → in_review → completed
| Status | Meaning |
|--------|---------|
| backlog | Extracted from blueprint, awaiting implementation plan |
| ready | Implementation plan attached, dependencies clear |
| in_progress | Developer is actively working |
| in_review | Code submitted for review; Delivery section populated with repo + PR |
| completed | Acceptance criteria verified; Delivery section finalized |
| blocked | Cannot proceed — blocker documented |
Terminal removal: WOs that are no longer needed are deleted (file removed, backlog entry removed). Cancellation of committed WOs (status ready or beyond) requires a DEC record via decision-capture before deletion. WOs that never left backlog can be deleted without a DEC. See DEC-031 #8 and #9.
File Placement
03-work-orders/
├── _backlog.md # Master index of ALL work orders (status-grouped)
├── intake/ # New WOs awaiting sprint triage
├── working/ # Graduated WOs (backlog, ready, in_progress, in_review)
└── done/ # Completed WOs
New work orders are always created in intake/. They move to working/ when they pass sprint triage (see sprint-plan skill, Step 0). They move to done/ when completed (see wo-update skill).
Commit Convention
work-orders: [action] [scope]
Examples:
work-orders: create WO-012 through WO-014 for password-resetwork-orders: update WO-012 status to in_progress
From the file to a first result
The download contains exactly the content displayed on this page. This copy may differ from the latest version in the source repository.
View the source and supporting filesThis download contains SKILL.md only. Any scripts, references, templates and dependencies are not included. Get the required files from the source, preserving their directory structure, before continuing.
1. Download the file
Download SKILL.mdFrom the root of a test project, save the file as .skill-downloads/create-work-orders-mm78nmqm/SKILL.md, or use the command below. This folder is for inspecting the skill before installation.
Command for macOS/Linux or WSL: requires a POSIX shell and curl. On Windows without WSL, use the download button and create the folder manually.
mkdir -p '.skill-downloads/create-work-orders-mm78nmqm' && curl --fail --show-error --location 'https://skills-guides.com/api/skills/create-work-orders-mm78nmqm/download' --output '.skill-downloads/create-work-orders-mm78nmqm/SKILL.md.part' && mv '.skill-downloads/create-work-orders-mm78nmqm/SKILL.md.part' '.skill-downloads/create-work-orders-mm78nmqm/SKILL.md'2. Check prerequisites
Read SKILL.md before using it. If it references scripts or external services, inspect them and prepare the required access in your test environment.
No recognized prerequisites section in this file. This does not mean it is self-contained: libraries, tools and access requirements may be mentioned elsewhere in the content or source repository.
The platforms on this page are catalog indications. The paths below follow tool documentation; they do not prove this skill has been run in those tools. No minimum tool version is confirmed for this listing.
3. Load the skill
After inspection, copy the complete folder into your project using one of the commands below, then open a new tool session. The command stops if a folder with the same name already exists: compare the files before replacing it.
Claude Code
test -f '.skill-downloads/create-work-orders-mm78nmqm/SKILL.md' && test -s '.skill-downloads/create-work-orders-mm78nmqm/SKILL.md' && mkdir -p '.claude/skills' && test ! -e '.claude/skills/wo-create' && cp -R '.skill-downloads/create-work-orders-mm78nmqm' '.claude/skills/wo-create'Cursor
test -f '.skill-downloads/create-work-orders-mm78nmqm/SKILL.md' && test -s '.skill-downloads/create-work-orders-mm78nmqm/SKILL.md' && mkdir -p '.cursor/skills' && test ! -e '.cursor/skills/wo-create' && cp -R '.skill-downloads/create-work-orders-mm78nmqm' '.cursor/skills/wo-create'OpenSkills
OpenSkills must already be installed. The local install/read procedure is checked with OpenSkills 1.5.0, which requires Node.js ≥ 20.6.0. Catalog slugs are not resolved automatically.
Official OpenSkills installationopenskills install './.skill-downloads/create-work-orders-mm78nmqm'openskills read 'create-work-orders-mm78nmqm'Reading should display the installed file content and its base directory. It does not run the skill or install its dependencies.
For an agent that reads AGENTS.md, then sync the index. This command changes AGENTS.md: review the diff before using it.
openskills syncManual method or another tool: attach SKILL.md to a conversation, or open it in an agent that can read files. Include any required supporting resources. No universal folder is recognized by every tool.
4. Try a first task
Use a short, synthetic input: an API endpoint, a small document or an interface brief. In a new conversation, attach the file or give the agent its path, then adapt this request:
Read .skill-downloads/create-work-orders-mm78nmqm/SKILL.md and apply its instructions to this example: [describe a test input]. Identify any missing files, dependencies or access. Produce the result requested by the skill without inventing data or running unreviewed scripts.Expected result: output matching the skill instructions, or a precise list of missing prerequisites. A read confirmation alone does not validate a first use. If an external resource is unavailable, that path remains untested.
React Hook Form Best Practices
Development
Master React Hook Form: avoid anti-patterns, manage subscriptions, and handle reset, dirty state, and controlled inputs correctly.
Emotion to Mantine Migration
Development
Migrate Emotion styled-components to Mantine components with style props and CSS modules. Use when converting .styled.tsx files or removing @emotion imports.
Scope Creep Detector
Development
Analyzes git diffs against a stated intent to detect scope creep, unrelated files, broad PRs, and oversized changes, then suggests keep, split, or justify actions.