estimate-performance-headroom

SkillDev tools

Lets your agent estimate how much performance improvement is left before hitting practical or theoretical limits.

Available today. Use it from your connected AI after setup.

Add ahel to your AI once: Claude, ChatGPT, Cursor, Claude Code or Codex. Then ask it to use this.

Then ask your AI: use the estimate-performance-headroom skill

About this skill

Estimate practical/human/theoretical ceilings and remaining performance headroom under stated assumptions.

What this skill tells your AI

The instructions your AI receives, as published by yogsoth-ai/de-anthropocentric-research-engine in skills/estimate-performance-headroom/SKILL.md and read by ahel’s review.

Purpose

Estimate practical, human, or theoretical ceilings and remaining performance headroom under stated assumptions.

Input contract

required: [current_performance, ceiling_reference, metric_schema]
optional: [human_baseline, oracle_bound, task_constraints, uncertainty_model]
constraints: [each ceiling must name its population, conditions, and assumptions]

Procedure

  1. Define the relevant ceiling and align its metric and conditions with current performance.
  2. Estimate the gap and uncertainty to each applicable ceiling.
  3. Separate attainable, theoretical, and assumption-dependent headroom.
  4. Identify evidence needed to reduce the dominant uncertainty.

Output contract

produces: [ceiling_estimates, headroom_estimates, assumption_register, uncertainty_priorities]
delta_fields: [findings, evidence_updates, uncertainties, recommended_jumps]

Quality gates

  • Ceiling and current score are comparable.
  • Headroom is not reported as an absolute fact when assumptions dominate.

Failure and counterexamples

Do not use an upper-bound theorem as a practical ceiling or equate benchmark maximum with human or task optimum.

Provenance map

  • resolved: headroom-estimation

Signals

GitHub stars
501
Forks
41
Last commit
Sep 2026
Advanced
Catalog kind
skill
Key
estimate-performance-headroom
Source
github.com/yogsoth-ai/de-anthropocentric-research-engine