Research Evidence Synthesizer

Research Gap Radar

Deterministically surface under-researched theme-by-method combinations from a paper corpus: build a theme-by-method coverage matrix, detect zero/low-coverage gaps (including combos that never co-occur), rank them, and flag under-studied themes. Byte-stable; no LLM. (Action Co-pilot: generates next-best-actions (actions.json), follow-up tasks and calendar reminders on top of the analysis — what to act on today.)

Citation Graph

Deterministically analyze a citation network from an edge list (citing -> cited): compute per-paper citation count (in-degree), references made (out-degree), and iterative PageRank (fixed damping/iterations), then rank the most-cited and most-influential papers. Byte-stable; no LLM. (Action Co-pilot: generates next-best-actions (actions.json), follow-up tasks and calendar reminders on top of the analysis — what to act on today.)

Reference Refinery

Citation format proofing: normalize references to GB/APA style. (Action Co-pilot: generates next-best-actions (actions.json), follow-up tasks and calendar reminders on top of the analysis — what to act on today.)

Literature Screen

Deterministically screen a paper list PRISMA-style: apply fixed-order inclusion/exclusion rules (year floor, minimum abstract length, required keyword, excluded keyword) to each row and produce PRISMA flow counts (identified -> per-stage survivors -> included / excluded-by-reason) with per-paper decisions. Byte-stable; no LLM. (Action Co-pilot: generates next-best-actions (actions.json), follow-up tasks and calendar reminders on top of the analysis — what to act on today.)

Literature Pilot

Deterministically project research-topic trends from a topic-by-year publication/citation table: estimate per-topic growth over a recent window, project forward under a scenario band, and rank emerging vs declining topics. Byte-stable; no LLM. (Action Co-pilot: generates next-best-actions (actions.json), follow-up tasks and calendar reminders on top of the analysis — what to act on today.)

Research Data Lens

Deterministic diagnostics for a research/experiment data table: descriptive statistics (EDA — per-column mean/median/SD/missing, IQR outliers, Pearson correlations) for dataset, or treatment-vs-control comparison with Cohen's d effect sizes for experiment. No LLM, byte-stable; results require human review. (Action Co-pilot: generates next-best-actions (actions.json), follow-up tasks and calendar reminders on top of the analysis — what to act on today.)