agentleFS
qingxuantangqingxuantang/Lennys-to-sop-and-skills/skills/decision-maker/SKILL.md

decision-maker

Skill16 starsChanged 10 months agoMITView on GitHub

Helps make high-quality decisions using proven frameworks from top experts. Use when: deciding between options, facing difficult choices, setting up decision processes, running pre-mortems, determining when to quit/continue, or needing group input on a decision. Includes: SPADE, Nominal Group, Kill Criteria, Pre-Mortem, Opportunity Cost frameworks. Sources: Gokul Rajaram, Annie Duke, Shreyas Doshi.

3 other projects use a version of this.

Adopted by

Most recently taken up by

  • Other versions3 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 3

    1. xcrrr/claude-skills★ 6same name
    2. molly-diversifiedfun/claude-skills★ 1same name
    3. tusshaarpd/skills★ 0same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository8 other files it gives its agents.

The file

Show its rules and text

We split this skill into the rules it gives an agent

25 rules. 0 are common, 25 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 3 edited versions we read keep these.

  • 25

    Distinctive

    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
  1. Distinctive:

    When decision ownership is unclear

    line 49
  2. Distinctive:

    Approvers: Must approve (keep minimal)

    line 61
  3. Distinctive:

    Create at least 3 alternatives (avoids false binary)

    line 66
  4. Distinctive:

    Always include "do nothing" as an option

    line 67
  5. Distinctive:

    Document decision and rationale

    line 75
  6. Distinctive:

    Only two alternatives (generate more)

    line 86
  7. Distinctive:

    When you need diverse input without groupthink

    line 100
  8. Distinctive:

    No interrupting, no "you're wrong"

    line 117
  9. Distinctive:

    Ask clarifying questions only

    line 118
  10. Distinctive:

    Don't force "alignment"—it doesn't exist

    line 123
  11. Distinctive:

    Check for signals regularly

    line 178
  12. Distinctive:

    When signal appears: execute pre-committed action

    line 179
  13. Distinctive:

    No debate—time for debate was when setting criteria

    line 180
  14. Distinctive:

    No action owner assigned

    line 195
  15. Distinctive:

    When stakes are high

    line 207
  16. Distinctive:

    When team seems overconfident

    line 208
  17. Distinctive:

    Don't share yet

    line 221
  18. Distinctive:

    Not making it independent (groupthink)

    line 243
  19. Distinctive:

    Not following up on mitigations

    line 245
  20. Distinctive:

    When ROI analysis feels incomplete

    line 259
  21. Distinctive:

    Don't ask: "Is this worth doing?" Ask: "Is this MORE worth doing than the best alternative?"

    line 274
  22. Distinctive:

    Ask: "Is this MORE worth doing than the best alternative?"

    line 275
  23. Distinctive:

    Only looking at ROI (ignores alternatives)

    line 300
  24. Distinctive:

    Not quantifying what you're giving up

    line 301
  25. Distinctive:

    Offer to combine frameworks: when appropriate (e.g., Pre-Mortem → Kill Criteria)

    line 320

“Common” is measured: kept by at least 6 in 10 of the 3 edited versions of this file we could read. “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

Decision-Maker Skill

Help users make better decisions using proven frameworks from Lenny's Podcast experts.

When This Skill Activates

  • "Help me decide..."
  • "Should we do A or B?"
  • "How do we make this decision?"
  • "We need alignment on..."
  • "Should we continue this project?"
  • "When should we quit?"
  • "Run a pre-mortem"
  • "I can't choose between..."
  • "What's the opportunity cost?"

Framework Selection Guide

SituationUse This Framework
Big strategic decision with stakeholdersSPADE
Need group input without groupthinkNominal Group
Deciding whether to continue/quitKill Criteria
Assessing risks before launchPre-Mortem
Comparing prioritization optionsOpportunity Cost

Framework 1: SPADE (Big Decisions)

Source: Gokul Rajaram - Lenny's Podcast Key Insight: Most decision delays come from unclear ownership and insufficient exploration of alternatives.

When to Use

  • Major product decisions
  • Strategic choices with significant tradeoffs
  • Cross-functional decisions requiring alignment
  • High-stakes, irreversible decisions
  • When decision ownership is unclear

Procedure

S - Setting (Define the context)

Template: "We need to decide [specific decision] by [date]
because [why it matters now]. Constraints: [list them]."

P - People (Assign roles)

  • Responsible: ONE decision-maker (not a committee)
  • Approvers: Must approve (keep minimal)
  • Consulted: Provide input
  • Informed: Told after decision

A - Alternatives (Generate options)

  • Create at least 3 alternatives (avoids false binary)
  • Always include "do nothing" as an option
  • Each must be viable with clear tradeoffs

D - Decide (Make the call)

  1. Gather input from Consulted parties
  2. Evaluate alternatives against criteria
  3. Decision-maker makes the call
  4. Get approval from Approvers
  5. Document decision and rationale

E - Explain (Communicate)

  • What was decided
  • Why this alternative was chosen
  • What alternatives were considered
  • What happens next

Pitfalls to Avoid

  • Vague decision framing
  • Multiple decision-makers (pick ONE)
  • Only two alternatives (generate more)
  • Relitigating after decision is made

Framework 2: Nominal Group Method (Group Decisions)

Source: Annie Duke - Lenny's Podcast Key Insight: The only thing that should happen in a meeting is discussion—discovery and decisions should happen independently.

When to Use

  • Brainstorming sessions
  • Forecasting and estimation
  • Any group decision-making process
  • When you need diverse input without groupthink

Procedure

Step 1: Discover Independently (Before Meeting)

  • Send question to all participants
  • Each person responds privately (no reply-all)
  • Collect: their answer + brief rationale

Step 2: Aggregate and Share

  • Compile all responses anonymously or attributed
  • Highlight the spread of opinions
  • Share before the meeting

Step 3: Discuss in Meeting (Focus on Disagreements)

  • 80% of time on differences, not agreements
  • Facilitator reflects back without opinion
  • No interrupting, no "you're wrong"
  • Ask clarifying questions only

Step 4: Decide Independently (After Meeting)

  • Single decision-maker decides, OR
  • Independent vote (can't see others' votes)
  • Don't force "alignment"—it doesn't exist

Step 5: Use "Nevertheless"

"I've heard everyone's input. Nevertheless, we're going
to do X. Here's why..."

Pitfalls to Avoid

  • Discovering opinions in the meeting (anchoring bias)
  • Letting loudest voice dominate
  • Seeking "alignment" (accept disagreement)
  • Senior person speaking first

Framework 3: Kill Criteria (When to Quit)

Source: Annie Duke - Lenny's Podcast Key Insight: Pre-mortems only work if paired with pre-committed actions you'll take when signals appear.

When to Use

  • Starting new projects or initiatives
  • Making investment decisions
  • Pursuing sales deals
  • Launching experiments
  • Any decision where you might need to quit later

Procedure

Step 1: Run Pre-Mortem

"Imagine it's [timeframe] from now and this has failed
completely. What were the early warning signals?"

Step 2: Categorize Signals

CategoryResponse
RedFatal—kill immediately
YellowSerious—requires intervention
OrangeConcerning—investigate

Step 3: Define Actions for Each Signal For each signal, specify:

  • What action to take (kill, pivot, investigate, escalate)
  • Who owns the action
  • Timeline to act

Step 4: Commit Before Launch

  • Review kill criteria with team
  • Get explicit agreement to act when signals appear
  • Document the commitments

Step 5: Monitor and Execute

  • Check for signals regularly
  • When signal appears: execute pre-committed action
  • No debate—time for debate was when setting criteria

Example Kill Criteria

Signal: Customer only discusses price, won't demo
Category: Red
Action: Kill the deal immediately
Owner: Account Executive
Reasoning: They're using us to beat up a competitor on price

Pitfalls to Avoid

  • Setting criteria but not committing to act
  • Renegotiating when signal appears ("but it's different now")
  • Vague signals that can't be observed
  • No action owner assigned

Framework 4: Pre-Mortem Technique (Risk Assessment)

Source: Shreyas Doshi - Lenny's Podcast Key Insight: Imagining failure before it happens surfaces risks that optimism blinds us to.

When to Use

  • Before launching projects
  • Before major initiatives
  • When stakes are high
  • When team seems overconfident

Procedure

Step 1: Set the Scene

"Imagine it's 6 months from now. This project has failed
completely. What went wrong?"

Step 2: Individual Brainstorm (5-10 min)

  • Each person writes 3-5 failure scenarios independently
  • Include specific, concrete reasons
  • Don't share yet

Step 3: Aggregate and Categorize

  • Collect all failure scenarios
  • Group similar ones
  • Identify themes

Step 4: Prioritize Top Risks

  • Vote on most likely/impactful
  • Focus on top 3-5 risks

Step 5: Create Mitigation Plans For each top risk:

  • Prevention: How to reduce likelihood
  • Detection: How to spot it early
  • Response: What to do if it happens

Step 6: Set Kill Criteria (Optional)

  • Convert key risks into kill criteria
  • Pre-commit to actions

Pitfalls to Avoid

  • Not making it independent (groupthink)
  • Stopping at identification (need mitigation)
  • Not following up on mitigations
  • Using it to justify not launching (that's not the point)

Framework 5: Opportunity Cost Mindset (Prioritization)

Source: Shreyas Doshi - Lenny's Podcast Key Insight: The true cost of doing something is everything else you can't do—not just the resources spent.

When to Use

  • Prioritization debates
  • Resource allocation decisions
  • "Should we do this?" discussions
  • When ROI analysis feels incomplete

Procedure

Step 1: Identify the Proposal

  • What's being proposed?
  • What resources does it require?
  • What's the expected outcome?

Step 2: List Alternatives (What You're NOT Doing)

  • What else could these resources accomplish?
  • What's the best alternative use?
  • Include "do nothing" and "do less"

Step 3: Compare Against Best Alternative Don't ask: "Is this worth doing?" Ask: "Is this MORE worth doing than the best alternative?"

Step 4: Factor in Hidden Costs

  • Context switching costs
  • Maintenance burden
  • Complexity added
  • Options foreclosed

Step 5: Factor in Hidden Benefits

  • Learning and optionality value
  • Platform/foundation value
  • Strategic positioning

Step 6: Make Explicit Tradeoff

"We're choosing to do X instead of Y because [reason].
We accept that Y won't happen."

The ROI Trap

  • ROI asks: "Does this return more than it costs?"
  • Problem: Everything with positive ROI looks good
  • Opportunity cost asks: "Is this the BEST use of resources?"

Pitfalls to Avoid

  • Only looking at ROI (ignores alternatives)
  • Not quantifying what you're giving up
  • Treating decisions as independent (they're not)
  • Ignoring option value

How to Apply This Skill

  1. Listen for the decision type
    • Big strategic → SPADE
    • Group input needed → Nominal Group
    • Continue/quit → Kill Criteria
    • Risk assessment → Pre-Mortem
    • Prioritization → Opportunity Cost
  2. Walk through the selected framework step-by-step with the user
  3. Highlight relevant pitfalls based on their situation
  4. Offer to combine frameworks when appropriate (e.g., Pre-Mortem → Kill Criteria)

Related Skills

  • /strategy-advisor - For strategic direction questions
  • /goal-setter - For goal-setting and prioritization
  • /leadership-coach - For team decision dynamics

Full SOPs (Deep Dives)

  • SPADE Framework
  • Nominal Group
  • Kill Criteria
  • Conviction-Based Persistence
  • Pre-Mortem
  • Opportunity Cost
  • Pivot vs Persevere
  • Pivot Decision Framework
  • Fear/Anger Decision

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.