Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Apply Affordance Theory (Gibson, 1979; Norman, 1988) to analyze the action possibilities that an artifact provides to an actor. Use this skill when the user needs to evaluate technology design from an affordance perspective, identify why users struggle with an interface, analyze IT-enabled organizational change through affordance actualization, or when they ask 'what does this technology afford', 'why can't users figure out this feature', or 'how does technology enable new practices'.
.claude/skills/asgard-ai-platform-grad-affordance/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-20 | ✗→✓ | ▲ Improved | 31% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 12% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 68% | 0% |
| case-10 | ✗→✓ | ▲ Improved | 17% | 0% |
| case-12 | ✗→✓ | ▲ Improved | 151% | 0% |
Affordance Theory explains how actors perceive and realize action possibilities offered by objects or environments. Gibson's ecological view holds that affordances are relational properties existing between actor and environment, independent of perception. Norman adapted the concept for design — perceived affordances guide user interaction, and signifiers communicate where action is possible. In IS research, affordance theory explains how technology enables (or constrains) organizational action.
IRON LAW: An affordance exists in the RELATION between actor and artifact —
it is neither a property of the object alone nor of the user alone.Key assumptions:
Specify the technology artifact and the actor groups. Characterize actor capabilities, goals, and context. The same artifact affords different things to a novice versus an expert.
For each actor-artifact pair, identify:
| Affordance Type | Description | |----------------|-------------| | Existence | What action possibilities objectively exist in the relation | | Perception | Which affordances actors actually perceive (Norman's focus) | | Actualization | Which perceived affordances actors act upon |
Include constraints (actions the artifact prevents) alongside affordances.
Identify three critical gaps:
For perception gaps: improve signifiers, onboarding, or documentation. For actualization gaps: provide training, resources, or remove organizational barriers. For false affordances: fix misleading cues in the interface.
markdown## Affordance Analysis: [Artifact] x [Actor Group] ### Actor Profile - Role: ... - Capabilities: ... - Goals: ... ### Affordance Map | Affordance | Exists? | Perceived? | Actualized? | Gap Type | |-----------|---------|------------|-------------|----------| | | | | | | ### Constraints - [constraint]: [effect on actor behavior] ### Gap Analysis - Perception gaps: ... - Actualization gaps: ... - False affordances: ... ### Recommendations 1. [Gap type]: [intervention] 2. ...
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-21 | pass→pass | 12,277 | 6,894 | -44% | 1 | 1 | 0% | 1,493 | 2,114 | +42% | 0 | 0 | — |
case-01 | fail→fail | 34,789 | 20,776 | -40% | 1 | 1 | 0% | 4,497 | 4,549 | +1% | 0 | 0 | — |
case-02 | fail→fail | 32,270 | 20,921 | -35% | 1 | 1 | 0% | 5,547 | 4,934 | -11% | 0 | 0 | — |
case-20 | fail→pass | 16,905 | 11,666 | -31% | 1 | 1 | 0% | 2,234 | 2,922 | +31% | 0 | 0 | — |
case-03 | fail→fail | 34,525 | 31,121 | -10% | 1 | 1 | 0% | 5,003 | 5,134 | +3% | 0 | 0 | — |
case-04 | pass→pass | 22,307 | 11,449 | -49% | 1 | 1 | 0% | 2,747 | 2,974 | +8% | 0 | 0 | — |
case-05 | fail→pass | 25,241 | 16,542 | -34% | 1 | 1 | 0% | 2,934 | 3,297 | +12% | 0 | 0 | — |
case-06 | fail→fail | 20,562 | 22,456 | +9% | 1 | 1 | 0% | 3,119 | 4,071 | +31% | 0 | 0 | — |
case-07 | pass→pass | 15,664 | 17,100 | +9% | 1 | 1 | 0% | 2,334 | 3,226 | +38% | 0 | 0 | — |
case-08 | pass→pass | 18,484 | 14,094 | -24% | 1 | 1 | 0% | 2,338 | 3,187 | +36% | 0 | 0 | — |
case-09 | fail→pass | 18,672 | 18,810 | +1% | 1 | 1 | 0% | 2,343 | 3,942 | +68% | 0 | 0 | — |
case-10 | fail→pass | 8,744 | 6,418 | -27% | 1 | 1 | 0% | 1,638 | 1,923 | +17% | 0 | 0 | — |
case-11 | fail→fail | 15,244 | 4,598 | -70% | 1 | 1 | 0% | 2,054 | 1,656 | -19% | 0 | 0 | — |
case-12 | fail→pass | 5,837 | 7,237 | +24% | 1 | 1 | 0% | 907 | 2,277 | +151% | 0 | 0 | — |
case-13 | pass→pass | 18,498 | 19,323 | +4% | 1 | 1 | 0% | 2,402 | 3,597 | +50% | 0 | 0 | — |
case-14 | pass→pass | 25,186 | 24,292 | -4% | 1 | 1 | 0% | 2,331 | 4,277 | +83% | 0 | 0 | — |
case-15 | pass→pass | 24,794 | 18,034 | -27% | 1 | 1 | 0% | 3,004 | 3,853 | +28% | 0 | 0 | — |
case-16 | fail→fail | 24,028 | 18,553 | -23% | 1 | 1 | 0% | 2,703 | 3,433 | +27% | 0 | 0 | — |
case-17 | pass→pass | 24,846 | 16,841 | -32% | 1 | 1 | 0% | 3,083 | 3,452 | +12% | 0 | 0 | — |
case-18 | pass→pass | 16,819 | 17,642 | +5% | 1 | 1 | 0% | 2,394 | 3,585 | +50% | 0 | 0 | — |
case-19 | pass→pass | 5,171 | 7,161 | +38% | 1 | 1 | 0% | 608 | 1,859 | +206% | 0 | 0 | — |
case-22 | pass→pass | 9,615 | 6,301 | -34% | 1 | 1 | 0% | 1,362 | 1,963 | +44% | 0 | 0 | — |
case-23 | pass→pass | 18,056 | 19,322 | +7% | 1 | 1 | 0% | 2,350 | 3,915 | +67% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 23 cases were attempted. The headline lift of +22 percentage points is the difference between those two pass rates over the 23 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.