← Back to Skills Marketplace
adawodu

Workflow Audit

by Adebayo Dawodu · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ✓ Security Clean
113
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install workflow-audit
Description
Conduct a structured operational audit — identify friction points, map workflows, quantify waste, and produce a priority-ranked automation blueprint with ROI...
README (SKILL.md)

Workflow Audit — The Friction X-Ray

Transform client conversations into a structured automation blueprint with hard ROI numbers. This is the deliverable for a diagnostic engagement.

When to Use

  • Client shares a meeting transcript describing their operations
  • Client describes manual processes, spreadsheet workflows, or tool-switching pain
  • You're conducting a discovery call and need structured output
  • Pre-sales: building the business case for a DynoClaw deployment

Input

The user provides one or more of:

  • Meeting transcript (Fathom, Otter, pasted text)
  • Description of current workflows and pain points
  • List of tools/systems in use
  • Team size and roles involved

If insufficient context, ask targeted discovery questions:

  1. "What's the ugliest spreadsheet your team updates every week?"
  2. "Where is your team copy-pasting data between two screens?"
  3. "What breaks when a key person goes on vacation?"
  4. "If you got 50 new clients tomorrow, which department collapses first?"
  5. "What does your team do on Saturday mornings that should've been done Friday?"

Processing Steps

Step 1 — Map the Current State

For each workflow mentioned, document:

  • Process name (e.g., "Weekly Executive Report", "Invoice Reconciliation")
  • Trigger — what starts this process
  • Steps — numbered sequence of actions
  • People involved — roles, not just names
  • Tools used — every system touched (CRM, spreadsheet, email, ERP, etc.)
  • Frequency — how often this runs
  • Time spent — estimated hours per occurrence
  • Error rate — how often it goes wrong and what the consequences are

Step 2 — Score Each Workflow (Friction Matrix)

Rate each workflow on two dimensions:

Volume (1-10): How often does this run?

  • 1-3: Monthly or less
  • 4-6: Weekly
  • 7-10: Daily or continuous

Cost of Error (1-10): What happens when it breaks?

  • 1-3: Minor inconvenience, easy to fix
  • 4-6: Delayed deliverables, client impact
  • 7-10: Financial loss, compliance risk, customer churn

Friction Score = Volume x Cost of Error (max 100)

Score Category Recommendation
70-100 High Volume, High Cost Automate immediately — highest ROI
40-69 Medium Strong automation candidate
15-39 Low-Medium Automate after high-priority items
1-14 Low Leave manual or defer indefinitely

Step 3 — Quantify the Waste

For each high-scoring workflow, calculate:

  • Hours/week wasted = (time per occurrence) x (frequency)
  • Annual cost = hours/week x 50 weeks x (blended hourly rate, default $45/hr for ops staff, $75/hr for senior staff)
  • Error cost = (error rate) x (cost per error) x (annual occurrences)
  • Total annual waste = labor cost + error cost

Step 4 — Design the Automation

For each workflow scoring 40+, propose:

  • Automation approach — which DynoClaw plugins/skills handle this
  • Data connections — what APIs or integrations are needed
  • Human checkpoints — where the human approves vs. where the agent acts autonomously
  • Failover plan — what happens when the AI encounters an edge case it can't handle
  • Implementation time — estimated hours to build and test

Step 5 — Build the ROI Model

Create a summary table:

Workflow Annual Waste Automation Cost Annual Savings ROI Payback
Process A $X $Y (one-time) + $Z/mo $W W/Y % N months

Total across all workflows:

  • Total annual waste identified
  • Total automation investment (one-time + annual)
  • Net annual savings
  • Payback period
  • 3-year cumulative savings

Step 6 — Produce the Blueprint

Format the final output as:

# Operational Automation Blueprint
**Prepared for:** [Client Name]
**Date:** [Date]
**Prepared by:** ParallelScore / DynoClaw

---

## Executive Summary
[2-3 sentences: We identified X workflows consuming Y hours/week at a cost of $Z/year. We recommend automating the top N workflows, yielding $W/year in savings with a payback period of M months.]

---

## Current State Assessment
[Workflow maps from Step 1]

## Friction Matrix
[Scored table from Step 2]

## Waste Quantification
[Dollar figures from Step 3]

## Recommended Automations
[Designs from Step 4, ordered by Friction Score descending]

## Investment & ROI Summary
[Table from Step 5]

## Implementation Roadmap
| Phase | Workflows | Timeline | Investment |
|-------|-----------|----------|------------|
| Phase 1 (Quick Wins) | Top 1-2 by score | Weeks 1-2 | $X |
| Phase 2 (Core Ops) | Next 2-3 | Weeks 3-6 | $Y |
| Phase 3 (Expansion) | Remaining | Month 2-3 | $Z |

## What We Need From You
- [List of accounts, access, contacts needed]

## Next Step
Reply with "Go" to begin Phase 1, or schedule a call to discuss.

Guidelines

  • Always quantify in dollars, not just hours — executives respond to money
  • Use conservative estimates — underpromise on savings so reality exceeds projections
  • Default labor rates: $45/hr ops staff, $75/hr senior staff, $150/hr executive time
  • If the client hasn't shared enough detail, ask the discovery questions — don't guess
  • The blueprint should be client-ready — professional enough to present to a CFO
  • Frame automation as "freeing your team for strategic work" not "replacing your team"
  • Always include human checkpoints — never propose fully autonomous workflows for the first engagement
  • Include a "What happens if you do nothing" section showing compounding cost of inaction
Usage Guidance
This skill is a reusable audit template and appears coherent and low-risk because it asks for no installs or credentials. Before using it, consider: 1) Provenance: there is no homepage or known publisher—if you plan to share sensitive customer data, verify who will see outputs and confirm the vendor/agent identity. 2) Data hygiene: do not paste secrets, passwords, or full credentials into transcripts; redact PII and account tokens. 3) When the blueprint asks for 'accounts, access, contacts needed', prefer least-privilege (read-only or scoped API tokens) and use temporary credentials or service accounts where possible. 4) Validate assumptions: confirm the default labor rates and error-cost assumptions with the client before presenting to finance. 5) If you need higher assurance, request a provenance/source and review a sample delivered blueprint on non-production data first.
Capability Assessment
Purpose & Capability
The name and description match the SKILL.md: it outlines steps to map workflows, score friction, quantify waste, design automations, and build an ROI model. It does not request unrelated binaries, env vars, or system access.
Instruction Scope
Runtime instructions are limited to: asking discovery questions, parsing supplied transcripts/descriptions, applying scoring and spreadsheet-like calculations, and formatting a deliverable. There are no instructions to read local files, access environment variables, or exfiltrate data beyond what the user pastes/provides.
Install Mechanism
No install spec and no code files — instruction-only skill. This minimizes persistence and on-disk risk.
Credentials
The skill declares no required environment variables, credentials, or config paths. Its stated data needs (meeting transcripts, workflow descriptions, team sizes) match the declared requirements.
Persistence & Privilege
always:false and user-invocable:true (normal). The skill does not request elevated privileges or modify other skills/config. Autonomous invocation is permitted by platform default but not requested by the skill beyond normal operation.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install workflow-audit
  3. After installation, invoke the skill by name or use /workflow-audit
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release — operational friction audit with ROI-driven automation blueprints
Metadata
Slug workflow-audit
Version 1.0.0
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 1
Frequently Asked Questions

What is Workflow Audit?

Conduct a structured operational audit — identify friction points, map workflows, quantify waste, and produce a priority-ranked automation blueprint with ROI... It is an AI Agent Skill for Claude Code / OpenClaw, with 113 downloads so far.

How do I install Workflow Audit?

Run "/install workflow-audit" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Workflow Audit free?

Yes, Workflow Audit is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does Workflow Audit support?

Workflow Audit is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Workflow Audit?

It is built and maintained by Adebayo Dawodu (@adawodu); the current version is v1.0.0.

💬 Comments