agentleFS
JustinTSmithJustinTSmith/ai-operator-skills/buyback-time-audit/SKILL.md

buyback-time-audit

Skill0 starsChanged 4 months agoMITView on GitHub

Run Dan Martell Buy Back Your Time Time and Energy Audit on the last 14 days of calendar. Use when the user asks for a time audit, buyback rate, DRIP matrix, Production Quadrant, Pain Line, Replacement Ladder, Camcorder Method, Perfect Week, or to measure creation vs coordination and delegate or automate low-value work.

The file

Show its rules and text

We split this skill into the rules it gives an agent

20 rules. About 0 are common, 16 are distinctive, and 4 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈16

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 4

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Run only Martell's Time and Energy Audit plus Buyback Loop.

    line 13
  2. Distinctive:

    Ask only for what is missing.

    line 28
  3. Distinctive:

    If a calendar skill is installed (google-calendar, caldav-calendar, accli), use it read-only

    line 33
  4. Distinctive:

    If the calendar is thin (empty blocks, "busy" with no title), say so and treat unlabelled hours as uncategorized, not as Production.

    line 38
  5. Distinctive:

    Do these in order. Do not skip scoring to jump to advice.

    line 42
  6. Distinctive:

    start, end, duration hours

    line 50
  7. Very specific:

    Write the working table to buyback-audit-YYYY-MM-DD.md in the user's workspace if you can write files.

    line 56
  8. Distinctive:

    Every row gets green (energizing / lights them up) or red (draining).

    line 60
  9. Distinctive:

    Do not use yellow. Ambiguous rows stay unscored and you ask.

    line 62
  10. Distinctive:

    Default mapping if they give no guidance

    line 82
  11. Distinctive:

    If they are at the Pain Line (calendar full, growth stalling, panic, "I should just sell"), say so plainly and keep the Transfer list short …

    line 118
  12. Very specific:

    Never invent meetings. After a successful pull, write raw JSON to ~/.openclaw/workspace/buyback-audits/raw-YYYY-MM-DD.json when the …

    line 136
  13. Very specific:

    If perfect-week is also installed, use it only in Fill (step 5), not as a substitute for the 14-day log.

    line 138
  14. Distinctive:

    Use assets/audit-report-template.md.

    line 142
  15. Distinctive:

    Create vs Coordinate heuristic

    line 152
  16. Distinctive:

    Create / Production - building, selling, writing, designing, coaching 1:1 that develops people

    line 154
  17. Distinctive:

    Do not recommend hiring before the 14-day table exists.

    line 160
  18. Distinctive:

    Do not fill freed time with more coordination.

    line 161
  19. Very specific:

    Do not treat "I like email" as Production if $ is low - that is Investment.

    line 162
  20. Distinctive:

    Do not dump the whole ceo-personal-os interview set.

    line 163

≈ “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

Buy Back Time Audit

Run only Martell's Time and Energy Audit plus Buyback Loop. Do not load Ferriss, Collins, Gerber, or other ceo-personal-os frameworks unless the user explicitly asks.

Tone - calm, direct, coach. No hustle slogans.

Credit - Buy Back Your Time by Dan Martell.

When this skill fires

  • time audit, time and energy audit, buy back my time
  • buyback rate, DRIP, Production Quadrant, Pain Line
  • calendar is full of status meetings / email / coordination
  • "what should I delegate" after looking at the last two weeks

Inputs to collect first

Ask only for what is missing. Do not stall if the user already pasted a calendar.

  1. Last 14 days of calendar (required)
    • Preferred - live Apple Calendar pull via scripts/pull_calendar_14d.sh
    • Fallback - ICS/CSV export or pasted agenda
    • If a calendar skill is installed (google-calendar, caldav-calendar, accli), use it read-only
  2. Annual income or income goal used for Buyback Rate. If refused, use a placeholder and label every $ rating as relative.
  3. Who can receive work - EA, partner, contractor, agent, nobody yet.
  4. Quarterly goals (optional but better). Used only in Fill.

If the calendar is thin (empty blocks, "busy" with no title), say so and treat unlabelled hours as uncategorized, not as Production.

Locked five-step procedure

Do these in order. Do not skip scoring to jump to advice.

1. Log the last 14 days

Chunk time into 15-minute blocks from the calendar. Merge contiguous identical events. Include evenings if they are on the calendar.

For each block record

  • start, end, duration hours
  • title / what actually happened
  • source (calendar title vs user correction)

Unscheduled work the user remembers (Slack, email triage, kid pickup) may be added as manual rows. Mark them source=manual.

Write the working table to buyback-audit-YYYY-MM-DD.md in the user's workspace if you can write files.

2. Score energy

Every row gets green (energizing / lights them up) or red (draining). If the user has not scored yet, propose a first pass from titles and ask them to correct reds they actually enjoy.

Do not use yellow. Ambiguous rows stay unscored and you ask.

3. Score dollar signs vs Buyback Rate

effective_hourly = annual_income / 2000
buyback_rate     = effective_hourly / 4

That is the same as annual_income / 8000. One buyback-rate hour is what you should pay someone else.

Rate each row $ to $$$$ relative to this person, not a generic market.

MarksMeaning
$At or below buyback rate. Someone else can do this cheaply.
$$Above buyback rate but not scarce. Trainable.
$$$High value, hard to hand off this quarter.
$$$$Only this person. Production or unique relationships.

Default mapping if they give no guidance

  • email triage, scheduling, inbox, expenses, travel booking, status standup, CRM hygiene → $
  • delivery ops, support, reporting, meeting notes → $ or $$
  • selling, closing, product bets, key-hire coaching, public content only they can do → $$$ or $$$$

4. Map DRIP and build the Transfer list

Plot each row on the DRIP matrix. Definitions live in references/drip-and-loop.md.

QuadrantMoneyEnergyAction
Delegationlow ($ / $$)redTransfer first
ReplacementhighredKeep until you hire/train a specialist, then transfer
InvestmentlowgreenCap the hours. Do not expand.
ProductionhighgreenProtect and Fill here

Transfer hit list = all Delegation rows, cheapest first. If there are no $ reds, take $$ reds.

For each hit-list item output

  • hours in the 14-day window and implied weekly hours
  • who should own it (EA / delivery / marketing / sales / leadership / agent)
  • Transfer method - Camcorder (record once, they write the SOP) vs checklist vs automate vs delete
  • 80% rule - ship when they hit ~80% of your quality, then 10-80-10 review

Replacement Ladder order when hiring is still the next step - Administration → Delivery → Marketing → Sales → Leadership. Do not recommend a Head of Sales while inbox and calendar are still on the founder.

5. Fill

Freed hours do not default to more email.

Fill with Production work tied to quarterly goals - selling, building the offer, coaching the team, content that compounds, rest if they are past the Pain Line.

Propose a Perfect Week sketch - deep-work blocks first, batched coordination in one or two windows, no status meetings that can be an async note.

If they are at the Pain Line (calendar full, growth stalling, panic, "I should just sell"), say so plainly and keep the Transfer list short enough to start this week.

Calendar ingest

On this Mac, pull Apple Calendar first. Read-only. Do not create or edit events.

${SKILL_DIR}/scripts/pull_calendar_14d.sh 14

That prints JSON { events: [{ start, end, hours, summary, calendar }] }. First run may prompt macOS Automation access for Calendar / Terminal / OpenClaw. If permission is denied, say so and stop.

Fallback order if Apple Calendar returns zero events

  1. python3 ${SKILL_DIR}/scripts/parse_ics_14d.py path/to/export.ics
  2. An installed calendar skill (google-calendar, caldav-calendar, accli, perfect-week)
  3. A pasted 14-day agenda

Never invent meetings. After a successful pull, write raw JSON to ~/.openclaw/workspace/buyback-audits/raw-YYYY-MM-DD.json when the workspace is writable.

If perfect-week is also installed, use it only in Fill (step 5), not as a substitute for the 14-day log.

Output format

Use assets/audit-report-template.md. Always include

  1. Buyback Rate math
  2. 14-day totals - hours by DRIP quadrant and by Create vs Coordinate vs Admin
  3. Leverage ratio - production_hours / coordination_hours (coordination = meetings + email + status + internal Slack-like blocks)
  4. Transfer list ranked by hours × drain
  5. What not to hire yet
  6. Fill plan for the next 7 days
  7. Open questions (unscored rows, missing income, missing delegate)

Create vs Coordinate heuristic

  • Create / Production - building, selling, writing, designing, coaching 1:1 that develops people
  • Coordinate - status meetings, standups, "quick sync", email processing, Slack catch-up, calendar Tetris
  • Admin - expenses, booking, tools, errands

Guardrails

  • Do not recommend hiring before the 14-day table exists.
  • Do not fill freed time with more coordination.
  • Do not treat "I like email" as Production if $ is low - that is Investment. Cap it.
  • Do not dump the whole ceo-personal-os interview set.
  • This is a time system, not therapy. If they describe panic or shutdown, name the Pain Line and keep the next action to one Transfer.

After the audit

Offer one follow-up only if useful

  • Camcorder script for the top Transfer item
  • Perfect Week calendar blocks
  • a 4-month reminder to run the loop again

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.