agentsclimarketplace

Agentd drug discovery

Skill rbr7/MedClawMini/skills/agentd-drug-discovery

A focused, production-minded library of 197 clinical-AI and healthcare data-science skills for the OpenClaw agent platform featuring data quality, clinical NLP, big-data ML, explainable AI, drug safety, and regulatory.

Install
npx -y skills add rbr7/MedClawMini --skill agentd-drug-discovery

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Use the AgentD workflow to mine evidence, design molecules, and rank candidates with SAR plus ADMET annotations for early drug discovery tasks.

SKILL.md

1.6 KB, as published. Nobody here has run it

At-a-Glance

  • description (10-20 chars): Hypothesis foundry
  • keywords: ligand-design, SAR, ADMET, docking, ranking
  • measurable_outcome: Generate ≥10 candidate molecules (or requested count) with SMILES, key properties, and rationales per run, all delivered within 15 minutes.

Inputs

  • target_protein, optional reference_compound, disease indication.
  • constraints dict (LogP, MW, TPSA, etc.) and num_candidates.

Outputs

  1. Ranked candidate list with SMILES + property scores + novelty metrics.
  2. ADMET/toxicity alerts and SAR rationale per molecule.
  3. Reproducibility manifest (data source versions, model checkpoints).

Workflow

  1. Evidence retrieval: Mine literature + databases for known ligands and liabilities.
  2. Generate candidates: Run AgentD generative step (scaffold hopping/fragment growth) aligned to constraints.
  3. Score & filter: Apply Lipinski/QED/ADMET heuristics; include docking setup when requested.
  4. Rank & explain: Combine efficacy, developability, novelty; summarize SAR learnings.
  5. Deliver outputs: Emit JSON/CSV plus narrative recommendations; mark as in silico.

Guardrails

  • Clearly state outputs are hypothetical and need wet-lab validation.
  • Flag PAINS/reactive motifs automatically.
  • Record data/model versions for audit trails.

References

  • Detailed parameter tables and dependencies listed in README.md.

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.