deanpeters/ai-product-operating-model-skills / skills/workflow-to-skill-converter
workflow-to-skill-converter
VerifiedConvert a proven, improved workflow into a governed, reusable skill with context, decisions, examples, guardrails, evaluations, ownership, and maintenance rules.
Workflow to Skill Converter
What Is It
Convert a workflow with demonstrated usefulness into a reusable skill that a human facilitator and an AI agent can run, inspect, evaluate, and improve. This is not a prompt-polishing exercise.
Why Use It
Local workflows disappear, drift, or spread without their judgment and failure modes. A skill can preserve the productive motion—but packaging an unproven workflow merely scales confusion.
When to Use It
Use after the workflow has a clear outcome, repeated examples, known decisions, and an owner. If the workflow is still unclear or ineffective, map and redesign it first.
What It Produces
- Reuse-readiness decision
- Canonical
SKILL.md, template, and examples - Context, guardrails, evaluations, and stop rules
- Stewardship, versioning, adoption, and improvement plan
Who Should Participate
Include workflow practitioners, the outcome owner, Product Operations or enablement, an agent-instruction author, and governance partners where consequences require them.
Evidence to Bring
Bring observed workflow examples, before-and-after evidence, decisions and exceptions, context inputs, failures, user feedback, and proof that the practice can be repeated.
How to Do It
- Confirm the workflow changes a useful decision or outcome.
- Assess reuse readiness: repeated use, stable core, known variation, owner, and evidence.
- Extract triggers, inputs, decisions, steps, outputs, handoffs, and completion criteria.
- Preserve judgment: options, tradeoffs, uncertainty, escalation, and stop rules.
- Separate essential instructions from templates, examples, references, and assets.
- Write trigger-rich metadata and concise imperative instructions.
- Add a worked and weak example that demonstrate reasoning.
- Define evaluation scenarios and forward-test without leaking expected answers.
- Assign stewardship, version, review cadence, adoption measures, and retirement rules.
Key Concepts
- Proven before packaged: reusable output requires evidence of a useful motion.
- Judgment preservation: encode decisions, not just steps.
- Progressive disclosure: keep core instructions lean; load details when needed.
- Stewardship: every reusable skill needs an owner and retirement path.
Organizational Applications
Use for discovery, synthesis, portfolio review, evidence review, governance, context assembly, and other recurring product-team motions.
Common Pitfalls
- Converting a blank template or one-off prompt
- Scaling a broken workflow
- Removing anti-patterns to shorten instructions
- Embedding confidential context
- Omitting evaluation, ownership, and retirement
- Measuring downloads instead of changed work or outcomes
Combine With
Use human-aipom-work-contract to define collaboration, aipom-workflow-playbook-builder for a deeper operating playbook, and aipom-adoption-impact-scorecard to measure changed practice.
Assets and Templates
Sources
This skill is an original AIPOM synthesis informed by the repository’s canonical skill specification and contribution workflow.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |