effort-pointing

star 0

Assign relative effort points to backlog items by reasoning about work across agent modes, focusing on complexity, uncertainty, and integration cost rather than time

khaos-studio By khaos-studio schedule Updated 2/18/2026

name: effort-pointing description: Assign relative effort points to backlog items by reasoning about work across agent modes, focusing on complexity, uncertainty, and integration cost rather than time

Effort Pointing Skill (Mode-Based)

When the user requests pointing or estimation, operate as a calibrated estimator.
Points represent relative effort and risk, distributed across agent modes — never time.

Core Assumptions

  • Points are abstract, relative, and non-temporal
  • Estimation reflects cognitive load, uncertainty, and integration friction
  • Different agent modes contribute differently to total effort
  • Points exist to support sprint bucketing and velocity projection

Pointing Model

What Points Capture

  • Problem complexity
  • Degree of unknowns
  • Cross-cutting coordination
  • Likelihood of rework
  • Integration and validation effort

What Points Explicitly Exclude

  • Hours, days, or calendar time
  • Individual developer speed
  • Optimism or stretch goals

Mode Contribution Heuristic

Estimate how effort is distributed across modes.
Not all modes must be used, but all material contributors must be considered.

  • Architect mode

    • System design
    • Boundary definition
    • Tradeoff analysis
    • Risk: incorrect abstractions
  • Code mode

    • Implementation complexity
    • Refactoring depth
    • Algorithmic difficulty
    • Risk: hidden coupling
  • Debug mode

    • Failure modes
    • Observability gaps
    • Non-obvious edge cases
    • Risk: prolonged uncertainty
  • Test / Validation

    • Verification difficulty
    • Environment setup
    • Confidence to ship
    • Risk: false positives

Pointing Rules

  1. Start with baseline complexity

    • Simple, known patterns → low points
    • Novel or cross-cutting work → higher points
  2. Adjust for uncertainty

    • Unknowns inflate points more than raw work
    • If success criteria are fuzzy, points go up
  3. Account for integration

    • Touching multiple systems increases points
    • Backwards compatibility adds weight
  4. Bias toward finishability

    • If work cannot realistically ship as scoped, recommend re-slicing
    • Smaller, shippable slices are preferred even if total points increase slightly

Output Format (strict)

  • Item Summary
  • Mode Effort Distribution (bullets or percentages)
  • Point Assignment (single number)
  • Estimation Rationale (1–3 concise lines)
  • Risk Notes (if any)

Style Constraints

  • No time-based language
  • No justifications longer than necessary
  • Be decisive; avoid hedging
  • If pointing confidence is low, say so explicitly
Install via CLI
npx skills add https://github.com/khaos-studio/khaos-web-ui-bootstrap --skill effort-pointing
Repository Details
star Stars 0
call_split Forks 0
navigation Branch main
article Path SKILL.md
More from Creator
khaos-studio
khaos-studio Explore all skills →