automate-audit
Skill3 starsChanged 7 months agoMITView on GitHub
Interview the user about their daily and weekly routines, then produce a ranked list of automation opportunities with effort-vs-impact scoring. Activates when the user asks "what should I automate?", "audit my workflow", "find my time wasters", or when they describe repetitive frustrations. Serves both technical and non-technical users.
We split this skill into the rules it gives an agent
7 rules. About 0 are common, 7 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈7
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 33
Ask one question at a time.
- Distinctive: line 37
Start broad: "Walk me through a typical workday from start to finish — what are the first few things you do when you sit down, and what …
- Distinctive: line 61
Technical users: Ask about CI/CD, deployments, code review, monitoring, testing
- Distinctive: line 62
Non-technical users: Ask about spreadsheets, email, scheduling, document creation, invoicing
- Distinctive: line 66
After interviewing, score each identified task on two dimensions:
- Distinctive: line 128
Don't oversell. If something genuinely requires human judgment and shouldn't be automated, say so and explain why.
- Distinctive: line 141
delegate: If the user isn't sure whether to automate or keep doing manually, use delegate to evaluate
≈ “Common” is an estimate: at least 1 in 5 of the 308,232 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.
Files
Automate Audit
Discover what the user is wasting time on that they don't realize could be automated. Interview them about their routines, then deliver a ranked automation opportunity report.
When to Activate
Manual triggers:
- "What should I automate?"
- "Audit my workflow"
- "Find my time wasters"
- "Where am I wasting time?"
- "Help me find things to automate"
Auto-detect triggers:
- User describes doing the same thing repeatedly ("every week I have to...", "I always end up manually...")
- User complains about a tedious process
- User mentions copy-pasting between tools, manual data entry, or repetitive formatting
Interview Phase
Ask one question at a time. Work through these areas in order, skipping what isn't relevant based on context.
1. Daily Rhythm
Start broad: "Walk me through a typical workday from start to finish — what are the first few things you do when you sit down, and what eats the most time?"
2. Weekly Recurring Tasks
"What do you do every week that feels like groundhog day? Reports, updates, check-ins, data pulls, scheduling?"
3. Tool Friction
"What tools or apps do you use daily? Where do you find yourself switching between tools or copying information from one place to another?"
4. The Dread Question
"What task do you actively dread or procrastinate on — not because it's hard, but because it's tedious?"
5. Communication Overhead
"How much time do you spend writing similar messages — status updates, follow-ups, onboarding emails, responses to common questions?"
6. Data and Reporting
"Do you regularly pull data, format it, or create reports? How manual is that process?"
Adapt your questions based on who you're talking to:
- Technical users: Ask about CI/CD, deployments, code review, monitoring, testing
- Non-technical users: Ask about spreadsheets, email, scheduling, document creation, invoicing
Analysis Phase
After interviewing, score each identified task on two dimensions:
Impact (1-5): How much time/frustration would automation save?
- 5 = Hours per week recovered
- 4 = Significant weekly time savings (1-2 hours)
- 3 = Moderate savings (30-60 min/week)
- 2 = Minor convenience
- 1 = Nice-to-have
Effort (1-5): How hard is it to automate?
- 1 = Can be done with a single Claude prompt or simple tool
- 2 = Needs a structured prompt or basic script
- 3 = Requires a workflow with multiple steps or tool integration
- 4 = Needs custom development or complex integration
- 5 = Requires significant engineering investment
Output: The Automation Opportunity Report
Present findings as a structured report:
# Automation Opportunity Report
## Quick Wins (High Impact, Low Effort)
These should be automated this week.
1. **[Task Name]** — Impact: X/5 | Effort: X/5
- What you're doing now: [current manual process]
- How to automate it: [specific recommendation]
- Tool/approach: [Claude prompt, Zapier, script, MCP server, etc.]
- Estimated time saved: [per week]
## High-Value Projects (High Impact, Higher Effort)
Worth investing in over the next month.
2. **[Task Name]** — Impact: X/5 | Effort: X/5
...
## Low-Priority (Lower Impact)
Automate these after the above are done.
3. **[Task Name]** — Impact: X/5 | Effort: X/5
...
## Summary
- Total tasks identified: X
- Quick wins: X (start here)
- Estimated weekly time savings if all quick wins are automated: X hoursRules
- Be specific in recommendations. Don't say "automate this" — say exactly how. Name the tool, write the prompt, describe the integration.
- For each quick win, offer to build the automation right now. "Want me to write the Claude prompt for this one?"
- Tailor technical depth to the user. If they're non-technical, recommend no-code tools (Zapier, Make, Claude prompts). If they're an engineer, suggest scripts, MCP servers, or Claude Code skills.
- Don't oversell. If something genuinely requires human judgment and shouldn't be automated, say so and explain why.
- The report should be actionable the same day — at least the quick wins. Don't produce a report that requires weeks of planning to act on.
Chaining
This skill pairs naturally with other Summit skills:
- decompose: For high-effort automation opportunities, use decompose to break them into manageable steps
- prompt-architect: For quick wins, use prompt-architect to build the Claude prompt that automates the task
- delegate: If the user isn't sure whether to automate or keep doing manually, use delegate to evaluate
In this repository98 other files it gives its agents.
- AGENTS.mdAGENTS.md
- agent-architectSkill
- architect-plan-for-dispatchSkill
- brain-dumpSkill
- chat-to-code-promptSkill
- cli-apifySkill
- cli-batSkill
- cli-duckdbSkill
- cli-fdSkill
- cli-ffmpegSkill
- cli-fzfSkill
- cli-ghSkill
- cli-httpieSkill
- cli-imagemagickSkill
- cli-jqSkill
- cli-lazygitSkill
- cli-pandocSkill
- cli-playwrightSkill
- cli-ripgrepSkill
- cli-sqliteSkill
- cli-yqSkill
- cli-yt-dlpSkill
- codebase-reviewSkill
- context-summarySkill
- decomposeSkill
- delegateSkill
- design-a-systemSkill
- design-doc-reviewSkill
- devils-advocateSkill
- gh-repo-createSkill
- ig-savesSkill
- interrogateSkill
- mentor-aristotleSkill
- mentor-attiaSkill
- mentor-aureliusSkill
- mentor-bloomSkill
- mentor-brownSkill
- mentor-buffettSkill
- mentor-builderSkill
- mentor-carnegieSkill
- mentor-clearSkill
- mentor-councilSkill
- mentor-cunninghamSkill
- mentor-dalioSkill
- mentor-davinciSkill
- mentor-demingSkill
- mentor-ferrissSkill
- mentor-feynmanSkill
- mentor-franklSkill
- mentor-garyveeSkill
- mentor-godinSkill
- mentor-gogginsSkill
- mentor-greeneSkill
- mentor-holidaySkill
- mentor-hormoziSkill
- mentor-houselSkill
- mentor-jockoSkill
- mentor-jungSkill
- mentor-kahnemanSkill
- mentor-kingSkill
- The first 60 shown.
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
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.

