Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Apply Giddens' structuration theory to analyze the duality of structure — how social structures are both the medium and outcome of the practices they organize. Use this skill when the user needs to bridge agency and structure in organizational or social analysis, explain how routines reproduce or transform institutional patterns, analyze the recursive relationship between action and structure, or when they ask 'do people shape institutions or do institutions shape people', 'how are these routine
.claude/skills/asgard-ai-platform-grad-structuration/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -19% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -23% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -2% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 3% | 0% |
| case-07 | ✗→✓ | ▲ Improved | -4% | 0% |
Structuration theory, developed by Anthony Giddens (1984), resolves the agency-structure dualism by proposing that structure has a "duality" — it is simultaneously the medium through which action occurs and the outcome that action produces. Structure does not exist independently of practice; it is instantiated in the moment of action and reproduced (or transformed) through ongoing practice.
IRON LAW: Structure does not exist independent of action — it is
PRODUCED and REPRODUCED through practice. Any analysis that treats
structure as a fixed, external constraint separate from human activity
violates the duality of structure.Key assumptions:
Select the recurring practice, routine, or pattern of interaction to analyze. Define the actors and the institutional context.
| Structural Dimension | Modality | Interaction Dimension | |---------------------|----------|----------------------| | Signification (meaning systems) | Interpretive schemes | Communication | | Domination (power relations) | Facility (resources) | Power | | Legitimation (normative rules) | Norms | Sanction |
For each structural dimension, identify:
Evaluate the degree to which agents exercise reflexive monitoring, and identify moments where routine reproduction gives way to structural transformation.
markdown## Structuration Analysis: [Context] ### Practice Under Analysis - Practice: [the recurring social practice] - Actors: [who performs it] - Institutional setting: [context] ### Structural Decomposition | Dimension | Structure (Rules/Resources) | Modality | Interaction | |-----------|---------------------------|----------|-------------| | Signification | [meaning systems in play] | [interpretive schemes] | [how actors communicate] | | Domination | [allocative and authoritative resources] | [facilities enabling power] | [how power is exercised] | | Legitimation | [norms and moral codes] | [normative expectations] | [how sanctions operate] | ### Duality of Structure | Dimension | Structure as Medium | Structure as Outcome | |-----------|--------------------|--------------------| | Signification | [how meaning enables action] | [how action reproduces/changes meaning] | | Domination | [how resources enable action] | [how action reproduces/changes power] | | Legitimation | [how norms enable action] | [how action reproduces/changes norms] | ### Agency Assessment - Discursive consciousness: [what actors can articulate about their practices] - Practical consciousness: [what actors know tacitly but cannot articulate] - Unintended consequences: [outcomes actors did not foresee] ### Reproduction vs. Transformation - Reproduction mechanisms: [how routines maintain structure] - Transformation potential: [where cracks, contradictions, or reflexive agency create change]
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 38,278 | 23,108 | -40% | 1 | 1 | 0% | 5,773 | 4,694 | -19% | 0 | 0 | — |
case-02 | fail→pass | 47,300 | 21,479 | -55% | 1 | 1 | 0% | 5,896 | 4,534 | -23% | 0 | 0 | — |
case-03 | fail→pass | 28,523 | 22,178 | -22% | 1 | 1 | 0% | 4,533 | 4,455 | -2% | 0 | 0 | — |
case-04 | fail→pass | 32,074 | 19,433 | -39% | 1 | 1 | 0% | 3,954 | 4,063 | +3% | 0 | 0 | — |
case-05 | fail→fail | 34,290 | 31,401 | -8% | 1 | 1 | 0% | 5,254 | 5,798 | +10% | 0 | 0 | — |
case-06 | fail→fail | 36,801 | 23,284 | -37% | 1 | 1 | 0% | 5,488 | 4,356 | -21% | 0 | 0 | — |
case-07 | fail→pass | 29,680 | 20,944 | -29% | 1 | 1 | 0% | 4,512 | 4,313 | -4% | 0 | 0 | — |
case-08 | fail→pass | 23,899 | 18,122 | -24% | 1 | 1 | 0% | 3,558 | 3,823 | +7% | 0 | 0 | — |
case-09 | fail→pass | 29,302 | 23,122 | -21% | 1 | 1 | 0% | 4,255 | 4,497 | +6% | 0 | 0 | — |
case-10 | fail→pass | 21,555 | 17,484 | -19% | 1 | 1 | 0% | 3,205 | 3,694 | +15% | 0 | 0 | — |
case-11 | pass→pass | 22,013 | 17,664 | -20% | 1 | 1 | 0% | 3,522 | 3,956 | +12% | 0 | 0 | — |
case-12 | fail→pass | 24,338 | 20,277 | -17% | 1 | 1 | 0% | 3,477 | 4,315 | +24% | 0 | 0 | — |
case-13 | fail→pass | 23,096 | 25,528 | +11% | 1 | 1 | 0% | 3,488 | 4,981 | +43% | 0 | 0 | — |
case-14 | pass→pass | 22,657 | 23,628 | +4% | 1 | 1 | 0% | 3,353 | 4,531 | +35% | 0 | 0 | — |
case-15 | fail→pass | 21,201 | 20,786 | -2% | 1 | 1 | 0% | 3,106 | 4,102 | +32% | 0 | 0 | — |
case-16 | fail→pass | 21,458 | 20,192 | -6% | 1 | 1 | 0% | 2,901 | 3,940 | +36% | 0 | 0 | — |
case-17 | pass→pass | 20,958 | 17,240 | -18% | 1 | 1 | 0% | 3,032 | 4,245 | +40% | 0 | 0 | — |
case-18 | fail→fail | 21,846 | 24,734 | +13% | 1 | 1 | 0% | 2,953 | 4,567 | +55% | 0 | 0 | — |
case-19 | pass→pass | 21,194 | 23,430 | +11% | 1 | 1 | 0% | 3,364 | 4,798 | +43% | 0 | 0 | — |
case-20 | pass→fail | 18,110 | 15,779 | -13% | 1 | 1 | 0% | 3,665 | 3,740 | +2% | 0 | 0 | — |
case-21 | pass→pass | 12,793 | 18,266 | +43% | 1 | 1 | 0% | 2,534 | 4,668 | +84% | 0 | 0 | — |
case-22 | pass→pass | 23,683 | 23,341 | -1% | 1 | 1 | 0% | 3,681 | 4,863 | +32% | 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. 22 cases were attempted. The headline lift of +50 percentage points is the difference between those two pass rates over the 22 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
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.