Kosmos

Research applicability

Documented multidisciplinary research workflow, with source-version and evaluation limits retained.

Getting started

Read the primary manuscript and its limitations first. Check implementation access and terms separately; this review did not run an installation, API call to a research system, or experiment.

Maintenance

Source snapshot checked 2026-10-19. Current availability and experimental version parity are separate claims.

License

Research code/data/manuscript terms require separate verification; repository metadata alone is not a reuse clearance.

Workflow & human oversight

L3 applies only to the researcher-objective literature and processed-data analysis workflow. A scientist supplies a dataset and research objective and evaluates/interprets the resulting report. No intermediate human interaction is claimed. The world model coordinates data-analysis/literature trajectories and revised hypotheses; current data-size/access constraints remain. Twenty-cycle examples and runs up to twelve hours do not establish correct findings. This assessment is limited to documented researcher-directed analysis, not general autonomous science or full independent replication.

Workflow level is not an evidence score; higher is not better evidence.

Approach

A structured world model coordinates literature and data-analysis agents, updates hypotheses over repeated cycles, and returns source-linked reports for scientist evaluation.

Outcomes (as reported by the originating team)

Seven showcased discoveries are not a representative success-rate denominator. Expert evaluators checked 102 statements from three reports: 79.4% supported overall; 85.5% data-analysis, 82.1% literature, 57.9% synthesis. Six-month-equivalent estimates are collaborator judgements, not controlled acceleration measurements.

Limitations

Study-recruited external statement checks are partial validation, not independent replication of the whole system. Direction sensitivity, roughly 5GB dataset limit and expensive human interpretation remain. Public trajectories/results do not establish full agent-code availability.

Reproducibility

Externally validated (partial), study-recruited statement checks; full independent reproduction not established. Documentary review only; no candidate code, model, simulation or experiment was executed. A public repository or result artifact is not proof of replication.

Adoption requirements

Untested suggestion: inspect the pinned sources and validate data, evaluator assumptions, implementation availability and researcher controls before planning a bounded trial. No setup or trial was executed.