ego-ability-ratio — quality + safety report
In the Skillier index (local__ego-ability-ratio) · scanned 2026-06-03 · engine: builtin+triage
✓ Clean — no heuristic safety flags surfaced.
Heuristic flags from the builtin scanner, which is known to over-flag (it trips on legitimate env-reading integrations, security skills, and library .eval calls). This is NOT an authoritative malicious verdict — re-scan with SkillSpector for the authoritative result. Run the authoritative scan →
📇 This skill is in the Skillier index (curated · deduped · quality-filtered). Install Skillier to route & load it into your AI client.
Quality notes
About this skill
Force a brutal ego-vs-ability diagnostic whenever someone's feedback loop to reality looks broken. Use this skill aggressively when the user is doing self-assessment, leader coaching, peer review of a senior person, 360 debrief, or asking "why won't this person listen", "the exec is defensive in…
📄 Read the SKILL.md
---
name: ego-ability-ratio
description: Force a brutal ego-vs-ability diagnostic whenever someone's feedback loop to reality looks broken. Use this skill aggressively when the user is doing self-assessment, leader coaching, peer review of a senior person, 360 debrief, or asking "why won't this person listen", "the exec is defensive in every review", "I keep getting the same complaints from my team", "am I the problem here", "I'm confident I'm right about this", "my ideas keep getting rejected and I don't know why", or "this person used to be great and now they're stuck". Also fires when a leader is surrounded by yes-people, when a smart person stops shipping, when seniority is being used to win arguments instead of evidence, or when a postmortem keeps externalizing blame. The diagnostic: if ego exceeds ability, the RL loop to reality is broken — name the gap, force responsibility internalization, restore the loop. Trigger eagerly even when the user does not name Musk or the framework.
---
# Ego Ability Ratio
> "A major failure mode is a high ego-to-ability ratio. If your ego-to-ability ratio gets too high, then you've broken the feedback loop to reality."
> — Elon Musk, *The Book of Elon* (Chapter: Frontline Leadership (ego subsection))
## What this skill captures
Musk frames ego as a literal control-systems problem, not a personality flaw. Every person — CEO, engineer, intern — runs a feedback loop between their actions and reality. Ego is the gain on the filter that lets reality's signal through. Crank ego too high relative to actual ability, and the filter starts rejecting any input that contradicts the self-image. "In AI terms, you'll break your reinforcement learning (RL) loop." The person stops learning, stops updating, and starts confidently failing in the same direction.
The discipline is therefore not "be humble" — it is operational: keep the ratio low enough that reality can still hit you. "Always be smashing your ego. Internalize responsibility... Do whatever the task is, whether grand or humble." The value you get from this skill: a concrete diagnostic that tells you whether the person in front of you (including yourself) still has a working feedback loop, plus the specific intervention to restore it before they ship more damage.
## When to use this skill
- A leader, founder, or senior IC is dismissing feedback and the team has stopped giving it.
- You are doing self-assessment and want to know if your confidence is calibrated or compensating.
- A previously high-performing person has plateaued, gone defensive, or started losing arguments by pulling rank.
- You are coaching someone whose stated ability does not match their shipped results.
- A postmortem keeps blaming external factors — vendors, market, team, "bad luck."
- A peer review reads as "they're talented but" and the team is quietly routing around them.
## The how-to
1. Estimate ability honestly, in shipped outcomes, not titles or self-report.
> "Whether you're a CEO or any other role, do whatever it takes to succeed."
> — *The Book of Elon*
Ability is what reality has actually confirmed: working systems, decisions that paid off, problems solved end-to-end. Strip the resume, the years, the credentials. What has this person — or you — actually produced in the last 6–12 months?
2. Estimate ego — the gap between self-image and that shipped ability.
> "A major failure mode is a high ego-to-ability ratio."
> — *The Book of Elon*
Ego is not confidence. Ego is the part of the self-model that is not load-bearing on results. Symptoms: certainty without evidence, rank-pulling in technical arguments, irritation at being corrected by juniors, narrating past wins to justify present positions.
3. Diagnose the feedback loop. Is reality still getting in?
> "If your ego-to-ability ratio gets too high, then you've broken the feedback loop to reality."
> — *The Book of Elon*
Concrete tests: When was the last time this person changed their mind based on data from someone junior? When did they last say "I was wrong"? Do their reports bring them bad news unprompted, or do they wait until it's too late? A broken loop is silent — the absence of correction is the symptom.
4. If the loop is broken, name it directly. Do not soften.
> "Always be smashing your ego. Internalize responsibility."
> — *The Book of Elon*
Soft framing ("you might want to consider...") feeds the ratio. The intervention has to be at the same magnitude as the distortion. Name the specific decisions where reality contradicted them and they kept going. If this is you — write the list yourself, before someone else has to.
5. Force responsibility internalization on a specific recent failure.
> "In AI terms, you'll break your reinforcement learning (RL) loop. You want to have a strong RL loop, which means internalizing responsibility and minimizing ego."
> — *The Book of Elon*
Pick one concrete miss. Have them — or yourself — say out loud: "I caused this. Here is the decision I made. Here is what I missed. Here is what I will do differently." If the sentence keeps drifting to "we" or "the team" or "the market," the loop is still broken. Run it again.
6. Make the humble task mandatory.
> "Do whatever the task is, whether grand or humble."
> — *The Book of Elon*
Ego survives on abstraction. It dies on contact with the actual work. Put the leader back in the codebase, the founder back on support tickets, the architect back on a bug fix. The point is not the work product — it is the recalibration of ability against reality.
## Common failure modes
- **Confusing humility-theater with a low ratio.** Saying "I might be wrong" while behaving as if you are not is still a broken loop. The diagnostic is behavioral, not verbal.
- **Letting tenure or title substitute for ability in the numerator.** "She's been here 12 years" is not ability. Ability is what shipped last quarter.
- **Treating this as a personality intervention.** Musk frames it as a control system. Fix the loop, not the person's character.
- **Surrounding the high-ratio person with people who won't push back.** This is how the ratio compounds. The org structure becomes the broken filter.
- **Doing this skill on others but never on yourself.** The person most likely to have a high ratio is the one running the diagnostic on everyone else.
## When NOT to use this skill
- Someone is genuinely high-ability and confident — that is calibration, not ego. Do not pathologize accuracy.
- Junior person who is overconfident but still in a tight feedback loop (frequent reviews, fast iteration). The loop will correct them; do not preempt it.
- The disagreement is actually a values or strategy difference, not an ability gap. Use a strategy framework, not this one.
- Acute crisis where the team needs decisive direction now — recalibrate egos after the fire is out, not during.
## Source
The Book of Elon by Eric Jorgenson (2026, Scribe Media). Chapter: "Frontline Leadership (ego subsection)" (in "What It Takes").Want a live grade + an embeddable README badge? Run your skill through the free scanner.
Graded independently by Skillproof — nothing to sell the author. Quality is mechanical + corpus-grounded; safety flags are heuristic (builtin+triage), not a malicious verdict.