Biomni · Agentic Research Directory

Biomni

Authors describe a general-purpose biomedical agent combining retrieval-augmented planning with code execution and biomedical tools; systematic benchmarking is author-reported.[2]

L3Multi-step tool-using agentic workflow Source Preprint (not peer-reviewed) Outcome Author-reported

Evidence & verification

Source level

Preprint (not peer-reviewed)
Grounded in a preprint that had not completed peer review at the time the source was checked.

Outcome verification

Author-reported
Outcomes are reported by the team that built the system. The entry is source-checked against the team's own publications; not independently reproduced.

Sources checked

2026-10-15 (UTC)

Author paper / prototype evidence source-checked (preprint version).[2] Benchmark results and generalization claims are author-reported and not independently reproduced.

Source-checked does not mean independently reproduced. See the Method page for how these labels are assigned.

Sources (1)

Research applicability

Authors describe a general-purpose biomedical agent combining retrieval-augmented planning with code execution and biomedical tools; systematic benchmarking is author-reported.[2]

Getting started

Start from the public repository or hosted interface; configure the agent environment and keep protected health information out of public tool usage, as the authors caution.

Maintenance

Maintenance status not independently audited.

License

Not verified; consult upstream terms and dependency/data/model licences before adoption; no independent license audit performed.

Workflow & human oversight

L3 reflects the authors' documented multi-step tool-using design; humans supervise and validate results, and reported outcomes remain author-reported.

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

Approach

An action-discovery step mines tools, databases and protocols from publications to build a unified agentic environment; a generalist agent then composes and carries out multi-step biomedical workflows using LLM reasoning, retrieval-augmented planning and code-based execution.[2]

Limitations

The retrieved article is explicitly a preprint version; reported benchmarks are author-reported and were not independently reproduced, and a later journal version was not used as validation. Authors warn the public tool is not for protected health information. No software execution for this review.

Reproducibility

Stored primary-source extract research-wide/candidate-evidence-003.md, accessed 2026-10-15, read for this review. No software, model or claimed outcome reproduced.

Adoption requirements

Configured agent environment, model access and appropriate biomedical datasets; expert oversight.

Comments

No comments yet.

Add a comment

Comments are held for moderation before appearing. Critical or corrective comments are welcome — especially evidence that an entry overstates what has been verified.