# Method Engine > > - **Type:** Skill - **Install:** `agentstack add skill-tobiasblask-open-paper-machine-method-engine` - **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior - **Seller:** [TobiasBlask](https://agentstack.voostack.com/s/tobiasblask) - **Installs:** 0 - **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills) - **Latest version:** 0.1.0 - **License:** MIT - **Upstream author:** [TobiasBlask](https://github.com/TobiasBlask) - **Source:** https://github.com/TobiasBlask/open-paper-machine/tree/main/skills/method-engine ## Install ```sh agentstack add skill-tobiasblask-open-paper-machine-method-engine ``` Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent. ## About > **Orchestration Log**: When this skill is activated, append a log entry to `outputs/orchestration_log.md`: > ``` > ### Skill Activation: Method Engine > **Timestamp:** [current date/time] > **Actor:** AI Agent (method-engine) > **Input:** [brief description of the methodology request] > **Output:** [brief description of what was produced — e.g., "DSR method section drafted with 3 evaluation criteria"] > ``` # Method Engine ## Method Selection Guide ### Decision Tree ``` What is your primary research goal? │ ├─ "I want to map what the literature says" │ → Systematic Literature Review (Section A) │ ├─ "I want to understand a phenomenon in depth" │ → Qualitative Study (Section B) │ ├─ Single context, deep → Single Case Study │ ├─ Multiple contexts, comparison → Multiple Case Study │ ├─ Build new theory from data → Grounded Theory / Gioia │ └─ Analyze text/documents systematically → Content Analysis (Mayring) │ ├─ "I want to test hypotheses / measure relationships" │ → Quantitative Study (Section C) │ ├─ Complex model with latent variables → SEM (PLS or CB) │ ├─ Simpler relationships → Regression │ └─ Experimental comparison → Experiment / RCT (Section F) │ ├─ "I want to build something (tool, framework, model)" │ → Design Science Research (Section D) │ ├─ "I want to combine approaches" │ → Mixed Methods (Section E) │ ├─ "I want to improve practice through iterative intervention" │ → Action Research (Section G) │ ├─ "I want to understand culture, practices, or lived experience" │ → Ethnography (Section H) │ ├─ "I want structured expert consensus on a complex issue" │ → Delphi Study (Section I) │ └─ "I want to model and test scenarios computationally" → Simulation (Section J) ``` --- ## Section A: Systematic Literature Review ### Method Section Template (ready to adapt): ``` 3. Research Methodology We conducted a systematic literature review following the guidelines of [vom Brocke et al. (2009, 2015) / Webster & Watson (2002) / Kitchenham & Charters (2007) / PRISMA 2020 (Page et al., 2021)]. This approach is appropriate because [justification: need to synthesize a growing but fragmented body of knowledge / field is maturing and needs stock-taking / practical guidance requires evidence synthesis]. 3.1 Search Strategy We searched [N] electronic databases: Semantic Scholar, OpenAlex, CrossRef, [and arXiv for preprints / and AIS eLibrary for IS-specific venues]. The search was conducted in [month/year] using the following query terms: [("generative AI" OR "generative artificial intelligence" OR "large language model*" OR "LLM" OR "GPT" OR "foundation model*") AND ("enterprise" OR "organization*" OR "business" OR "implementation" OR "adoption")] [("AI agent*" OR "autonomous agent*" OR "agentic AI") AND ("organization*" OR "enterprise" OR "business process" OR "implementation")] The search was limited to publications from [year] to [year], in [English / English and German]. 3.2 Selection Criteria Table [N] summarizes our inclusion and exclusion criteria. | ID | Criterion | Rationale | |----|-----------|-----------| | IC1 | Peer-reviewed journal article or conference paper | Quality assurance | | IC2 | Focuses on [topic] in organizational context | Scope alignment | | IC3 | Published between [year] and [year] | Recency | | IC4 | Available in English [or German] | Accessibility | | EC1 | Purely technical (no organizational dimension) | Out of scope | | EC2 | Editorial, book review, or abstract-only | Insufficient depth | | EC3 | Duplicate publication | Avoid double-counting | 3.3 Search and Screening Process Figure [N] presents the PRISMA flow diagram of our search and selection process. The initial search yielded [N] records across all databases. After removing [N] duplicates, [N] records were screened based on title and abstract, of which [N] were excluded. The remaining [N] articles were assessed in full text, resulting in [N] studies included in the final synthesis. [Forward and backward citation tracking (snowballing) on the [N] most-cited included studies identified an additional [N] relevant papers, bringing the total to [N] studies.] 3.4 Data Extraction and Analysis From each included study, we extracted: [list categories: research question, theoretical lens, methodology, sample/context, key findings, limitations, and contribution type]. We synthesized findings using a concept-centric approach (Webster & Watson, 2002), organizing results in a concept matrix that maps studies against key themes identified through iterative reading and coding. ``` ### PRISMA Flow Diagram (text version): ``` Identification: Records from Semantic Scholar: [n] Records from OpenAlex: [n] Records from CrossRef: [n] Records from arXiv: [n] Records from manual/snowballing: [n] ───────────────────────────────── Total identified: [N] Duplicates removed: -[n] Records after deduplication: [N] Screening: Title/abstract screened: [N] Excluded: -[n] Full-text assessed: [N] Eligibility: Full-text excluded (with reasons): -[n] - Not organizational context: [n] - Not empirical/conceptual: [n] - Not accessible: [n] Included: Studies in final synthesis: [N] ``` --- ## Section B: Qualitative Methods ### Case Study (Yin, 2018 / Eisenhardt, 1989) Method section template: ``` 3. Research Methodology We employed a [single/multiple] case study approach (Yin, 2018) to investigate [phenomenon] in [context]. Case study research is appropriate when investigating a contemporary phenomenon within its real-world context, particularly when the boundaries between phenomenon and context are not clearly evident (Yin, 2018). 3.1 Case Selection [For single case:] We selected [case] as a [revelatory/critical/typical/extreme] case (Yin, 2018) because [justification]. [For multiple cases:] Following [theoretical/literal] replication logic (Yin, 2018), we selected [N] cases based on [selection criteria]. Table [N] provides an overview of the cases. | Case | Industry | Size | AI Maturity | Selection Rationale | |------|----------|------|-------------|---------------------| | A | [X] | [X] | [X] | [X] | | B | [X] | [X] | [X] | [X] | 3.2 Data Collection We collected data from multiple sources to enable triangulation (Yin, 2018): - [N] semi-structured interviews with [roles] (average duration: [X] minutes) - Internal documents: [list types] - [Observation / workshop protocols / system logs] - [Archival data: annual reports, press releases] All interviews were recorded and transcribed [verbatim / in summary form]. 3.3 Data Analysis We analyzed the data using [thematic analysis (Braun & Clarke, 2006) / the Gioia methodology (Gioia et al., 2013) / qualitative content analysis (Mayring, 2014)]. [Method-specific description — see subsections below.] 3.4 Research Quality We ensured research quality through: - **Construct validity**: Multiple data sources, chain of evidence - **Internal validity**: Pattern matching, explanation building - **External validity**: [Replication logic across cases / analytical generalization] - **Reliability**: Case study protocol, case study database ``` ### Gioia Methodology (Gioia et al., 2013) ``` Data analysis followed the Gioia methodology (Gioia et al., 2013). First, we engaged in open coding of interview transcripts and documents, identifying first-order concepts that remained close to informant language. This yielded [N] initial codes. Through constant comparison and iterative abstraction, we grouped these into [N] second-order themes reflecting more abstract, researcher-driven categories. Finally, we aggregated themes into [N] overarching dimensions that form the basis of our emerging framework. Figure [N] presents the resulting data structure. ``` ### Mayring Content Analysis (Mayring, 2014) ``` We analyzed the data using qualitative content analysis following Mayring (2014). We employed [deductive / inductive / mixed] category formation. [Deductive:] Categories were derived from [Theory/prior framework] and applied to the material. Coding rules and anchor examples were defined a priori. [Inductive:] Categories emerged from the data through systematic paraphrasing, generalization, and reduction of text passages. After coding [X]% of the material, the category system was revised and finalized. [Both:] Inter-coder reliability was assessed using [Cohen's κ / Krippendorff's α], yielding a value of [X], indicating [substantial/excellent] agreement. ``` --- ## Section C: Quantitative Methods ### Survey + PLS-SEM ``` 3. Research Methodology 3.1 Research Model and Hypotheses [Refer to Section 2 where hypotheses were developed] 3.2 Measurement All constructs were measured using validated scales from prior literature. [Construct 1] was measured with [N] items adapted from [Author] (Year). [Construct 2] used [N] items from [Author] (Year). All items were assessed on a [7-point Likert / 5-point Likert] scale. Table [N] lists all measurement items with their sources. 3.3 Data Collection Data were collected via an online survey distributed to [target population] through [distribution channels] between [month] and [month year]. After removing incomplete and inattentive responses (attention check items, completion time [X] / IQR ≤ [X] / ≥ [X]% agreement] were considered to have reached consensus. [Optional: Round 4 if consensus not reached] 3.3 Consensus Measurement We assessed consensus using [Kendall's W coefficient of concordance / interquartile range (IQR) / percentage agreement / coefficient of variation]. Consensus was defined as [specific threshold, e.g., IQR ≤ 1 on a 7-point scale / Kendall's W > 0.7 / ≥ 70% agreement]. 3.4 Analysis Final results were analyzed by [computing mean/median rankings, identifying clusters of related items, comparing across expert subgroups]. [If applicable:] Non-consensus items were analyzed qualitatively to understand divergent views. ``` ### Panel Size Guidelines: - Minimum: 10-15 experts (Okoli & Pawlowski, 2004) - Typical: 15-30 experts - Attrition: Plan for 20-30% dropout per round --- ## Section J: Simulation / Agent-Based Modeling ### Method Section Template: ``` 3. Research Methodology We employ a [system dynamics / agent-based / discrete event / Monte Carlo] simulation approach following [Law (2015) / Gilbert & Troitzsch (2005) / Sterman (2000)]. Simulation is appropriate because [the phenomenon involves complex dynamic interactions that are difficult to study empirically / we need to explore scenarios and parameter sensitivities / ethical or practical constraints prevent real-world experimentation]. 3.1 Model Design The simulation model represents [system/phenomenon] with the following key components: **Agents/Entities:** | Agent Type | Attributes | Behavior Rules | Count | |------------|-----------|----------------|-------| | [Type 1] | [list] | [decision rules] | [N] | | [Type 2] | [list] | [decision rules] | [N] | **Environment:** [Describe the simulation environment: topology, resources, constraints] **Interaction Rules:** [Describe how agents interact: communication, competition, cooperation] **Time:** The model runs in [discrete time steps / continuous time] over [N iterations / time horizon]. 3.2 Theoretical Grounding The model's behavioral rules are grounded in [theory/empirical findings]: - [Rule 1] is based on [Author (Year)] who found [finding] - [Rule 2] reflects [theoretical mechanism from Theory X] - [Parameter values] were calibrated using [empirical data / expert estimates / literature values] 3.3 Implementation The model was implemented in [NetLogo / AnyLogic / Python (Mesa) / Matlab / Vensim / R] (version [X]). [Key implementation choices and simplifying assumptions]. The source code is available at [repository URL]. 3.4 Verification and Validation Following Sargent (2013), we conducted: - **Verification** (does the model run correctly?): [code review, debugging, unit tests, trace analysis, comparison with analytical solutions] - **Validation** (does the model represent reality?): [comparison with empirical data, face validation by [N] domain experts, sensitivity analysis, extreme condition tests] 3.5 Experimental Design We explored [N] scenarios varying [parameters]: | Scenario | Parameter 1 | Parameter 2 | Rationale | |----------|------------|------------|-----------| | Baseline | [value] | [value] | Reference case | | S1 | [value] | [value] | Test [hypothesis/what-if] | | S2 | [value] | [value] | Test [hypothesis/what-if] | Each scenario was run [N] times (Monte Carlo replications) to account for stochastic variation. Results are reported as [mean ± SD / median with 95% CI / distribution plots]. ``` --- ## Research Data Management (RDM) ### FAIR Principles (Wilkinson et al., 2016) Every research project should address data management. Use this section as a checklist and to draft the data availability statement. | Principle | Requirement | How to Address | |-----------|-------------|----------------| | **Findable** | Data has persistent identifier, rich metadata | DOI via Zenodo/Figshare, descriptive README | | **Accessible** | Data retrievable via standardized protocol | Open repository, clear access conditions | | **Interoperable** | Data uses shared vocabularies/formats | Standard formats (CSV, JSON, BibTeX), codebooks | | **Reusable** | Data has clear license, provenance | CC-BY 4.0, data collection documentation | ### Data Management Plan Template ```markdown ## Data Management Plan ### 1. Data Description - **Type:** [survey responses / interview transcripts / system logs / simulation output / literature database / code] - **Format:** [CSV / JSON / PDF / audio / text] - **Volume:** [estimated size] - **Sensitivity:** [public / restricted / confidential] ### 2. Data Collection - **Method:** [how data will be collected] - **Tools:** [instruments, platforms, software] - **Timeline:** [when collection starts and ends] ### 3. Documentation and Metadata - **Codebook:** [variable descriptions, coding schemes] - **README:** [project overview, file structure, usage instructions] - **Provenance:** [data sources, transformation steps, version history] ### 4. Ethics and Legal Compliance - **Consent:** [informed consent process, template reference] - **Anonymization:** [strategy: pseudonymization, k-anonymity, aggregation] - **GDPR/DSGVO:** [legal basis, data protection measures, DPO contact] - **Ethics approval:** [IRB/ethics board reference number] ### 5. Storage and Backup - **Active storage:** [institutional server / cloud / local — with encryption] - **Backup:** [3-2-1 rule: 3 copies, 2 media types, 1 offsite] - **Retention:** [how long after project completion] ### 6. Sharing and Archiving - **Repository:** [Zenodo / Figshare / institutional / discipline-specific] - **License:** [CC-BY 4.0 / CC0 / restricted access with justification] - **Embargo:** [if applicable: duration and reason] - **DOI:** [will be assigned upon deposit] ### 7. Responsibilities - **Data steward:** [name/role] - **Access control:** [who can access what, and how] ``` ### Data Availability Statement Templates ``` [Open data:] The data that support the findings of this study are openly available in [repository] at https://doi.org/[DOI], reference number [ref]. [Restricted data:] The data that support the findings of this study are available from [source/organization] but restrictions apply to the availability of these data, which were used under license for the current study, and so are not publicly available. Data are however available from the authors upon reasonable request and with permission of [source/organization]. [No data (theor … ## Source & license This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code. - **Author:** [TobiasBlask](https://github.com/TobiasBlask) - **Source:** [TobiasBlask/open-paper-machine](https://github.com/TobiasBlask/open-paper-machine) - **License:** MIT Install and usage instructions live in the source repository linked above. ## Pricing - **Free** — Free ## Security capabilities Automated source analysis of v0.1.0 — what this tool can access: - **Network access:** no - **Filesystem access:** no - **Shell / process execution:** no - **Environment & secrets:** no - **Dynamic code execution:** no *"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.* ## Versions - **0.1.0** — security scan: passed — Imported from the upstream source. ## Links - Listing page: https://agentstack.voostack.com/l/skill-tobiasblask-open-paper-machine-method-engine - Seller: https://agentstack.voostack.com/s/tobiasblask - Browse the marketplace: https://agentstack.voostack.com/browse --- Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.