TobiasBlask

AUTONOMOUS PAPER PRODUCTION AGENT. This is the primary agent of the plugin. Takes a paper title, topic, or research question and AUTONOMOUSLY executes the complete research-to-draft pipeline. The user acts as orchestrator — approving direction at checkpoints, not doing the work. Activate for ANY request to write, start, create, or draft a paper. Also activate when the user provides a paper title or research topic. NEW in v6.3: Phase 0 (Idea Evaluation) — before producing, evaluates whether the paper is worth writing using 7 dimensions, the conclusion-first test, and RS1-RS8 research strategy principles.

Activate when the user needs to select, justify, describe, or execute a research methodology. Provides method selection guidance, complete method section templates, quality criteria, and tool recommendations. Covers SLR, qualitative (case study, Gioia, Mayring, Grounded Theory), quantitative (SEM, regression, survey, experiment), DSR, mixed methods, action research, ethnography, Delphi study, and simulation. Also includes Research Data Management (RDM) guidance for FAIR-compliant data handling.

Activate when the user wants to simulate a double-blind peer review of their paper before submission or before sharing with co-authors. Reads the current draft (draft.md or paper.tex), generates 2 independent reviewer reports in the style of top IS/CS conferences (ICIS, ECIS, MISQ level), and saves the output as simulated_reviews.md. The output is formatted to serve as direct input for /respond-reviewers (review-engine feedback loop).

Activate when the user wants to audit a paper's empirical or technical claims against a linked code repository — checking whether experiments, datasets, models, metrics, and hyperparameters described in the paper actually exist and match the code. Produces a structured audit report classifying each claim as CONFIRMED, PARTIAL, MISSING, or MISMATCH, with file/line evidence. Useful for reproducibility checks, reviewer due diligence, and pre-submission self-audits of ML/CS/empirical papers that ship code.

Activate when the user needs to evaluate whether a research idea is worth pursuing, brainstorm new research directions, or stress-test a paper concept before committing. This is Phase 0 of the paper machine — the gate that decides whether the full 8-phase pipeline should run. Also activates standalone via /evaluate-idea. Integrates Carlini's research philosophy (conclusion-first test, taste for problems, kill conditions, unreasonable effort) with structured evaluation (7 dimensions, RS1-RS8 principles, 3 specialist agents).

ALWAYS activate when the user needs to find, organize, review, or synthesize academic literature. Uses academic APIs (Semantic Scholar, OpenAlex, CrossRef, arXiv) via scripts/academic_search.py. Handles search strategy, snowballing, screening, concept matrices, narrative synthesis, and literature monitoring (detecting new publications since last search). NEVER use web scraping for paper discovery — APIs first, web search only for verification.

Activate when the user wants to export a completed paper draft to production-ready LaTeX (.tex) and PDF. Converts draft.md + references.bib + figures/ into a complete arxiv-style LaTeX project with properly resolved \citep/\citet citations, booktabs tables, figure environments, and compiled PDF output.

Activate when the user needs to analyze how their paper positions itself relative to the closest existing work. Builds a differentiation matrix, identifies unique positioning, and generates a positioning statement that strengthens the contribution argumentation. Complements the theory-engine by focusing on the competitive landscape of related papers.

Activate when the user needs to create conference presentation slides from a completed paper. Extracts key content, designs a slide structure, generates slide content with speaker notes, and produces a presentation-ready markdown file. Supports IS/CS conference formats (15-20 min presentations).

Activate when the user wants to prepare a paper for submission to a specific venue. Handles venue-specific formatting validation, anonymization checks for double-blind review, cover letter generation, suggested reviewer identification, and submission checklist completion. Produces a submission-ready package.

Activate when the user needs to systematically screen papers for a Systematic Literature Review (SLR). Implements the PRISMA-compliant screening pipeline: define inclusion/exclusion criteria, title/abstract screening, full-text screening, quality assessment, and PRISMA flow diagram generation. Takes the literature_base.csv from Phase 1 (Reconnaissance) and produces a filtered, documented, auditable set of included studies.

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.

ALWAYS activate when the user writes, drafts, or revises any part of an academic paper. This is the core skill for overcoming writer's block and producing complete first drafts. Provides concrete sentence-level templates, paragraph formulas, and section blueprints for IS/WI/BWL research papers. Works for journal papers (MISQ, BISE, EJIS), conference papers (ICIS, ECIS, WI), and working papers.

Strategic project triage — should you continue, pivot, or kill your current research project? Evaluates 5 signals (results, competition, impact, effort, motivation) and delivers a clear Continue/Pivot/Kill recommendation. For Pivot: suggests concrete new framing. For Kill: suggests what to salvage.

Activate when the user needs to generate, refine, or evaluate academic figures, diagrams, or statistical plots. Uses PaperBanana to transform text descriptions or data files into publication-quality illustrations via direct Python API call. Fallback: matplotlib/seaborn.

Activate when the user needs to manage multi-author collaboration on a paper. Tracks author contributions using the CRediT taxonomy, manages responsibility assignments, documents the human-AI division of labor, and produces an author contribution statement ready for submission.

Write one specific section of a paper as complete paragraphs. Use when you already have context and just need one section drafted. Follows writing-engine templates.

Generate conference presentation slides from a completed paper. Extracts key content, builds a slide structure, generates content with speaker notes, and produces a Marp-compatible markdown slide deck.

Simulate a double-blind peer review of the current paper. Generates 2 independent reviewer reports in the style of top IS/CS conferences (ICIS, ECIS, MISQ level). Each report includes Summary, Strengths, Weaknesses, Minor Comments, and Overall Recommendation. Output is saved as simulated_reviews.md and can be fed directly into /respond-reviewers for systematic revision.

Search academic literature across Semantic Scholar, OpenAlex, CrossRef, and arXiv. Runs Phase 1 (Reconnaissance) of the paper machine as a standalone operation.

Evaluate whether a research idea is worth pursuing before investing months of effort. Stress-tests ideas along 7 dimensions (novelty, impact, timing, feasibility, competition, nugget clarity, narrative potential) using 3 specialist agents. Delivers a PURSUE / PARK / KILL verdict with actionable next steps. Based on Carlini's research philosophy and the RS1-RS8 principles.

Export the current paper draft to production-ready LaTeX (.tex) and PDF using the arxiv-style template. Converts draft.md + references.bib + figures/ into a complete LaTeX project with resolved citations, tables, and figures. Compiles to PDF automatically.

Generate an academic diagram from a text description using PaperBanana. Provide a section of your paper or a methodology description and get a publication-quality figure.

Creative research brainstormer that generates ideas researchers working within a single field would miss — cross-field connections, challenges to conventional wisdom, and reframings. Produces bold, specific, surprising ideas rather than safe incremental ones. Activate during Phase 0 (DIVERGE phase) of the paper machine pipeline, when the user asks for research ideas, or when exploring new directions.

START THE PAPER MACHINE. Give a title, topic, or research question — the autonomous agent takes over and runs the complete 5-phase pipeline: Reconnaissance → Framing → Architecture → Production → Assembly. You approve at checkpoints, the machine does the work.

Activate when the user needs to verify citations, check source accuracy, or validate that referenced papers actually support the claims attributed to them. Retrieves source abstracts and full text (for open-access papers), compares each citation claim against actual source content, and produces a structured verification report with classifications (VERIFIED, PLAUSIBLE, MISMATCH, UNVERIFIABLE, NOT FOUND).

Verify all citations by fetching source abstracts/full-text and checking claims against actual content

Systematically screen papers for a Systematic Literature Review (SLR). Applies PRISMA-compliant title/abstract and full-text screening with documented inclusion/exclusion criteria. Takes literature_base.csv from Phase 1 and produces a filtered, auditable set of included studies with PRISMA flow diagram.

Prepare a paper for submission to a specific venue. Runs anonymization checks for double-blind review, generates a cover letter, identifies suggested reviewers, validates venue-specific formatting, and produces a complete submission checklist. Creates a submission-ready package.

Adversarial but constructive research sparring partner that stress-tests research ideas BEFORE the researcher invests months of effort. Evaluates ideas along 7 dimensions, delivers a Pursue / Refine / Kill verdict, and identifies the single most important question to resolve next. Activate when evaluating research ideas, when the user asks "is this worth pursuing?", "should I write this paper?", or during Phase 0 of the paper machine pipeline.

Senior research advisor for project-level strategic decisions. Handles the harder questions: Should I keep going? Am I working on the right thing? What am I giving up? Operates in 5 modes: Project Triage, Comparative Advantage Mapping, Impact Forecasting, Opportunity Cost Analysis, and Scooping Risk Assessment. Activate when the user asks strategic research questions, during Phase 0 of the paper machine pipeline (Phases 4-5: DEEPEN and DECIDE), or when evaluating whether to continue/pivot/kill an ongoing project.

Activate when the user needs to analyze qualitative data — interview transcripts, field notes, or open-ended survey responses. Handles structured summarization, thematic coding, cross-case analysis, theme matrices, and evidence retrieval. Designed to solve the context-window problem: generates compact summaries first, then works from summaries instead of full transcripts. Only loads full text when specific quotes are needed. Supports Gioia, Mayring, Grounded Theory, and general thematic analysis workflows.

Activate when the user provides reviewer or co-author feedback (annotated PDF, pasted comments, or reviewer report) and wants to implement revisions. Extracts review points, maps them to paper.tex locations, classifies actions, implements changes, recompiles, and generates a change log + latexdiff. Handles the full revision loop from feedback to committed changes.

Analyze qualitative interview transcripts. Generates structured summaries, thematic codes, cross-case analysis, and evidence tables. Solves context-window overflow by working summary-first.

Analyze how the paper positions itself relative to the closest existing work. Identifies the 5-10 most similar papers, builds a differentiation matrix, and generates a positioning statement that strengthens the contribution argumentation.

Analyze academic writing quality of the current draft. Checks for passive voice overuse, weak openings, hedging balance, readability, sentence length variation, and word repetitions. Produces a writing quality report with specific improvement suggestions per section.

Audit a paper's empirical claims against the linked code repository — catches MISMATCH, MISSING, and PARTIAL claims before submission

Creative research brainstorming — generate ideas researchers within a single field would miss. Cross-field connections, assumption challenges, alternative framings, and wild cards. Produces bold, specific, surprising ideas rather than safe incremental ones.

Generate a statistical plot from CSV/JSON data using PaperBanana. Provide a data file and describe the visualization intent.

Re-run literature search queries to find newly published papers since the last search. Compares new results against the existing literature_base.csv and highlights papers that appeared after the initial search date.

Process reviewer or co-author feedback and implement revisions. Accepts annotated PDF, pasted comments, or reviewer report. Extracts review points, classifies actions, presents change plan for approval, implements changes, recompiles LaTeX, generates latexdiff, and documents everything.

Scooping risk assessment — is someone else working on your research idea? Identifies competing groups, assesses pace of publication, evaluates whether the core insight is "in the air", and provides a watch list with search terms, key researchers, and venues to monitor.