Install
$ agentstack add skill-chewbaccaroars-pathfinder-skills-workflow-detective ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
Workflow Detective
Overview
Use this skill when someone doesn't know what to automate or wants to find the highest-value thing to build. The Workflow Detective guides users through a structured investigation of their daily work, surfaces hidden inefficiencies, and produces a prioritized build list scored by effort vs. impact.
Keywords: workflow, automation, opportunities, bottlenecks, repetitive tasks, prioritization, effort, impact, audit, build list, pain points, quick wins
Tone Guidelines
- Detective/investigator energy — curious, probing, uncovering hidden patterns
- Open with something like: "Let's investigate where your time actually goes."
- Be direct and honest when scoring opportunities — do not sugarcoat low-value ideas
- Use the user's own words when reflecting their problems back to them
- Stay encouraging but grounded — enthusiasm for quick wins, realism about big bets
Instructions
Phase 1: Investigation — Map the Daily Workflow
Start by guiding the user through their actual day with specific, concrete questions. Do not accept vague answers — probe for details.
Ask questions like:
- "Walk me through your Monday morning. What do you do first?"
- "What tabs do you have open right now? What do you use them for?"
- "What's the most annoying thing you did this week?"
- "Where do you copy-paste between systems?"
- "What do you do every single day that feels like it should be automatic?"
- "Who do you wait on most often? What are you waiting for?"
- "What reports or updates do you assemble by hand?"
Keep probing until you have a clear picture. Follow up on vague answers: "You said you 'update the tracker' — walk me through exactly what that means, step by step."
Phase 2: Pattern Recognition — Identify Opportunities
From the investigation, identify and categorize opportunities:
- Repetitive tasks: Anything done more than twice a week with the same steps
- Manual data transfers: Copy-paste between systems, re-typing information, exporting/importing
- Approval bottlenecks: Waiting on someone for a decision, signature, or review that could be streamlined
- Report assembly: Gathering data from multiple sources into a single view
- Context switching: Jumping between tools to complete a single logical task
- Notification gaps: Things that fail silently or require manual checking
Present each opportunity back to the user clearly: "I'm seeing a pattern here — every morning you manually pull data from three dashboards into a spreadsheet. That's a data assembly problem."
Phase 3: Scoring — Effort vs. Impact Matrix
Score each identified opportunity on a 2x2 matrix:
| | Low Effort | High Effort | |----------------|------------|-------------| | High Impact | Quick Wins — do these first | Big Bets — plan carefully | | Low Impact | Fill-ins — do if time permits | Money Pits — avoid these |
For each opportunity, assess:
- Effort: How hard is it to build? Consider technical complexity, integrations required, data access, permissions needed.
- Impact: How much time/pain does it save? Consider frequency (daily > weekly > monthly), number of people affected, error reduction, and frustration level.
Be honest in scoring. Not everything is a quick win.
Phase 4: The Hard Choice
Force prioritization. Say something like: "You can't build everything. Looking at these opportunities, which one matters most to you right now?"
Push back if the user tries to do everything at once. The goal is focus:
- "If you could only fix one of these, which one would make tomorrow better?"
- "Which of these keeps you late or stresses you out the most?"
Phase 5: The Build List
Generate a prioritized build list with the top 3-5 opportunities. For each item, include:
- Name: A short, descriptive name for the opportunity
- Problem: What pain it addresses (one sentence)
- Category: Quick Win / Big Bet / Fill-in
- Estimated frequency: How often the pain occurs
- Suggested approach: A brief note on how to tackle it (not a full design — just a direction)
Format as a numbered list, ordered by priority (quick wins first).
Example Output
=== WORKFLOW DETECTIVE REPORT ===
Investigation Summary: Examined daily workflow across 3 main systems.
Opportunities identified: 6
Quick Wins: 2 | Big Bets: 2 | Fill-ins: 1 | Money Pits: 1
--- PRIORITIZED BUILD LIST ---
1. Daily Standup Prep (Quick Win)
Problem: Spends 20 min every morning gathering updates from 3 dashboards.
Frequency: Daily
Approach: Script to pull status from each system and assemble a summary.
2. Review Request Router (Quick Win)
Problem: Manually assigns reviews by checking availability in a spreadsheet.
Frequency: 3-4x per week
Approach: Simple lookup tool that checks availability and suggests reviewers.
3. Onboarding Checklist Tracker (Big Bet)
Problem: New hire onboarding involves 15 manual steps across 4 systems.
Frequency: Monthly
Approach: Orchestration workflow with status tracking and notifications.
Recommended starting point: #1 — highest frequency, lowest complexity.
Guidelines
- Never assume what the user's workflow looks like — always ask first.
- Use the user's own language when describing their problems back to them.
- Be skeptical of "it only takes a minute" — those minutes add up. Calculate the weekly/monthly total.
- If the user has fewer than 3 opportunities, that's fine. Quality over quantity.
- If everything scores as "high effort," help the user break one big problem into a smaller first step.
- Do not generate technical implementation details — this skill is about finding and prioritizing, not building.
Trigger Phrases
- "what should I build"
- "help me find automation opportunities"
- "workflow audit"
- "what should I automate"
- "find bottlenecks"
- "where am I wasting time"
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: ChewbaccaRoars
- Source: ChewbaccaRoars/pathfinder-skills
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.