theory-engine

name	theory-engine
description	Activate when the user needs to select a theoretical lens, formulate a research gap, derive hypotheses or design principles, or write a contribution statement. Provides concrete theory-to-paper templates, not abstract advice.

Orchestration Log: When this skill is activated, append a log entry to outputs/orchestration_log.md:

### Skill Activation: Theory Engine
**Timestamp:** [current date/time]
**Actor:** AI Agent (theory-engine)
**Input:** [brief description of the theory selection/gap formulation request]
**Output:** [brief description of what was produced — e.g., "Selected sociotechnical systems theory, drafted theoretical framing (800 words)"]

Theory Engine

Purpose

Turn "I don't know which theory to use" into a concrete theoretical framing with drafted paragraphs ready for the paper.

Step 1: Match Topic to Theory

Theory Selection by Research Topic

AI Adoption & Implementation:

Theory	Use When	Key Constructs	Seminal Reference
TOE Framework	Studying organizational adoption factors	Technology, Organization, Environment	Tornatzky & Fleischer (1990)
UTAUT/UTAUT2	Studying individual user acceptance	Performance/Effort Expectancy, Social Influence	Venkatesh et al. (2003/2012)
Absorptive Capacity	Studying how orgs learn/integrate new AI knowledge	Acquisition, Assimilation, Transformation, Exploitation	Cohen & Levinthal (1990)
Dynamic Capabilities	Studying how orgs adapt to AI disruption	Sensing, Seizing, Reconfiguring	Teece (2007)
Socio-Technical Systems	Studying joint optimization of social + technical	Social subsystem, Technical subsystem, Joint optimization	Bostrom & Heinen (1977)
Institutional Theory	Studying mimetic/coercive/normative adoption pressures	Isomorphic pressures, Legitimacy	DiMaggio & Powell (1983)

AI Agents & Automation:

Theory	Use When	Key Constructs	Seminal Reference
Agency Theory	Studying delegation to autonomous agents	Principal, Agent, Information asymmetry	Eisenhardt (1989)
Affordance Theory	Studying what AI enables in practice	Affordances, Constraints, Actualization	Markus & Silver (2008)
Human-AI Collaboration	Studying augmentation vs. replacement	Complementarity, Task allocation	Raisch & Krakowski (2021)
Paradox Theory	Studying tensions from AI adoption	Control-autonomy, Transparency-efficiency	Smith & Lewis (2011)

Digital Transformation (broadly):

Theory	Use When	Key Constructs	Seminal Reference
Ambidexterity	Studying exploration/exploitation balance	Exploitation, Exploration, Structural/Contextual	O'Reilly & Tushman (2013)
Resource-Based View	Studying AI as strategic resource	VRIN resources, Competitive advantage	Barney (1991)
Organizational Learning	Studying how orgs learn from AI experiments	Single/double-loop learning	Argyris & Schön (1978)
Structuration Theory	Studying technology-in-use patterns	Structure, Agency, Duality	Giddens (1984) / Orlikowski (2000)

For the GenAI/Agents Paper Specifically

Most promising theoretical lenses:

1. Socio-Technical Systems Theory — GenAI/Agents change both the technical AND social system; implementation success depends on joint optimization
2. Dynamic Capabilities — Organizations need to sense AI opportunities, seize them through implementation strategies, and reconfigure processes
3. Affordance Theory — GenAI/Agents offer specific affordances (text generation, autonomous decision-making) that are actualized differently depending on organizational context
4. Paradox Theory — AI agents create inherent tensions (autonomy vs. control, efficiency vs. transparency, innovation vs. risk)

Step 2: Formulate the Gap

Gap Formula Templates

Template A — Fragmented Knowledge:

While prior research has examined [aspect 1] (Author, Year; Author, Year) and 
[aspect 2] (Author, Year; Author, Year) in isolation, an integrative understanding 
of [how these aspects interact / the full picture] is lacking. This is problematic 
because [specific consequence of the fragmentation for theory or practice].

Template B — New Phenomenon, Existing Theory Untested:

[Theory] has proven valuable for understanding [prior phenomenon] (Author, Year). 
However, the emergence of [new phenomenon, e.g., autonomous AI agents] introduces 
dynamics that [Theory] has not yet been applied to — specifically, [what's new: 
e.g., agent autonomy, non-deterministic outputs, emergent behavior]. Whether and 
how [Theory]'s core mechanisms operate in this new context remains an open question.

Template C — Practice Outpacing Theory:

Organizations are rapidly adopting [technology/practice], as evidenced by 
[practitioner evidence: industry reports, adoption statistics]. Yet academic 
research lags behind: existing studies [are mostly conceptual / focus on 
narrow aspects / rely on early-stage data]. There is an urgent need for 
[systematic empirical investigation / theoretically grounded guidance / 
comprehensive frameworks] to inform both scholarship and practice.

Template D — Methodological Gap:

The existing body of research on [topic] is dominated by [dominant method: 
surveys / case studies / conceptual papers]. This methodological concentration 
limits our understanding because [what it can't capture: longitudinal dynamics / 
cross-contextual comparison / causal mechanisms / implementation details]. 
A [different method] approach can complement existing knowledge by [what it adds].

For the GenAI/Agents Paper:

Likely combination of B + C:

While research on AI adoption in organizations has a rich tradition grounded in 
theories such as [TOE/UTAUT/Dynamic Capabilities], the emergence of generative AI 
and autonomous AI agents introduces qualitatively new dynamics — including 
non-deterministic outputs, emergent agent behavior, and the delegation of cognitive 
tasks previously considered uniquely human. These characteristics challenge 
fundamental assumptions in existing adoption and implementation frameworks. 
Meanwhile, organizational practice is advancing rapidly, with [X]% of enterprises 
piloting GenAI solutions (Source, Year), yet academic research providing systematic, 
theoretically grounded implementation guidance remains scarce.

Step 3: Derive Hypotheses or Propositions

For Quantitative Studies (Hypothesis Derivation):

Template:

[Theory] suggests that [theoretical mechanism] (Author, Year). In the context 
of [your study], this implies that [contextual application], because [reasoning 
linking theory to context]. Therefore:

H[N]: [Independent variable] [positively/negatively] influences [dependent variable] 
      [in the context of / when...].

For Qualitative / SLR Studies (Proposition or Research Framework):

Template:

Based on [Theory] and the reviewed literature, we propose that [relationship or 
dynamic]. This proposition integrates insights from [Author1] (Year), who showed 
[finding], with [Author2] (Year), who demonstrated [finding]. We expect that 
[expected pattern], which we investigate through [method].

For DSR (Design Principles):

Template (Chandra Kruse et al., 2016):

DP[N]: For [artifact type] to [aim/purpose] in [context], 
       [actors/the system] should [action/feature] 
       because [justification from kernel theory] (Theory, Author, Year).

Step 4: Write the Contribution Statement

The 3-Contribution Formula (most robust):

This study makes three contributions. First, we contribute to the literature on 
[stream 1] by providing [specific deliverable: a systematic overview / an empirical 
test / a design artifact / an integrative framework]. While prior work has [what they 
did], we [what's new and different]. 

Second, we [theoretical contribution]. By applying [Theory] to [new context/phenomenon], 
we [extend/challenge/refine] its explanatory scope. Specifically, we show that 
[concrete theoretical insight].

Third, we offer [practical contribution] in the form of [actionable deliverable: 
guidelines / a taxonomy / implementation recommendations / a decision framework]. 
This enables [practitioners/managers/organizations] to [specific benefit].

Contribution Type Self-Check

Before finalizing, verify your contribution is one of these (not "Routine Application"):

Type	Test	Example
New knowledge	Does this tell us something we didn't know?	First systematic review of GenAI agent implementation
New lens	Does this apply theory in a genuinely new way?	Agency theory for human-AI delegation
New method	Does this introduce a methodological innovation?	Framework for evaluating agent autonomy levels
New artifact	Does this create something usable? (DSR)	Implementation maturity model for GenAI
Integration	Does this connect previously separate ideas?	Linking adoption factors + org change + governance