agentleFS
Sign inStart free
jpeslar1jpeslar1/john-peslar-ai-skills/ai-workflow-architect/SKILL.md

ai-workflow-architect

Skill12 starsChanged 4 months agoMITView on GitHub

Design practical AI workflows, agents, automations, SOPs, and human approval loops for business operations. Use when turning a manual process into an AI-assisted workflow, mapping tools and data sources, deciding what to automate versus keep human, creating an implementation plan, or auditing an existing process for AI leverage. Trigger on "automate this", "AI workflow", "AI agent", "SOP", "manual process", "operations automation", "Claude workflow", "Codex workflow", or "where can AI save time?

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

6 rules. About 0 are common, 5 are distinctive, and 1 is 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.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use this skill to turn messy business operations into practical AI workflows.

    line 12
  2. Very specific:

    Read references/workflow-map.md when the user wants a full workflow map, SOP, or implementation spec.

    line 14
  3. Distinctive:

    Do not automate: unclear, high-risk, or low-volume

    line 35
  4. Distinctive:

    Start with the smallest workflow that proves value.

    line 84
  5. Distinctive:

    Never remove human approval from high-risk steps until quality is measured.

    line 85
  6. Distinctive:

    Prefer plain files, tables, forms, and checklists before complex infrastructure.

    line 86

≈ “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

AI Workflow Architect

Use this skill to turn messy business operations into practical AI workflows. Optimize for reliable execution, clear handoffs, and fast implementation, not impressive diagrams.

Read references/workflow-map.md when the user wants a full workflow map, SOP, or implementation spec.

Core Principle

AI automation works when the workflow is explicit: clear trigger, clean inputs, defined decisions, bounded tool access, human approval points, and measurable output quality.

Workflow

  1. Map the current process
    • Trigger
    • Inputs
    • Tools
    • Human decisions
    • Outputs
    • Failure points
    • Time spent
  2. Classify each step
    • Automate: repetitive, rule-based, low-risk
    • Assist: judgment-heavy but draftable
    • Human approval: customer-facing, financial, legal, reputational, or irreversible
    • Do not automate: unclear, high-risk, or low-volume
  3. Design the AI workflow
    • Agent role
    • Data sources
    • Tool permissions
    • Prompts or skills needed
    • Memory or knowledge base needed
    • Human review gates
    • Logging and audit trail
  4. Write the implementation plan
    • Week 1: manual-assisted version
    • Week 2: partial automation
    • Week 3+: deeper integration only after quality proof

Output Format

**Current Workflow**
[summary]

**Automation Map**
| Step | Current Owner | AI Role | Human Gate | Risk |
|---|---|---|---|---|

**Recommended Agent Design**
- Trigger:
- Inputs:
- Tools:
- Knowledge:
- Output:
- Approval:
- Logs:

**Implementation Plan**
1. ...
2. ...
3. ...

**Risks**
- ...

**First Test**
[smallest useful workflow to run this week]

Rules

  • Start with the smallest workflow that proves value.
  • Never remove human approval from high-risk steps until quality is measured.
  • Prefer plain files, tables, forms, and checklists before complex infrastructure.
  • Name what can break.
  • Include observability: logs, status, error handling, and review queues.

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.

Check your setup, free