Scores drug targets for discovery campaigns with GO/NO-GO decisions, evidence trails from Open Targets/ChEMBL/PDB/ClinicalTrials.gov, and structured reports.
From clawbionpx claudepluginhub clawbio/clawbio --plugin clawbioThis skill uses the workspace's default tool permissions.
checksums.sha256commands.shdemo_input.jsonenvironment.ymltarget_validation_scorer.pytests/test_target_validation_scorer.pyProvides UI/UX resources: 50+ styles, color palettes, font pairings, guidelines, charts for web/mobile across React, Next.js, Vue, Svelte, Tailwind, React Native, Flutter. Aids planning, building, reviewing interfaces.
Fetches up-to-date documentation from Context7 for libraries and frameworks like React, Next.js, Prisma. Use for setup questions, API references, and code examples.
Calculates TAM/SAM/SOM using top-down, bottom-up, and value theory methodologies for market sizing, revenue estimation, and startup validation.
You are Target Validation Scorer, a specialised ClawBio skill for drug discovery. Your role is to score therapeutic targets across 5 evidence dimensions and return a transparent GO/NO-GO decision.
This is not a prediction tool. It is a decision support tool that makes the reasoning behind target selection transparent and reproducible.
Typical use case: prioritising targets for early-stage drug discovery campaigns before committing computational or experimental resources.
output_directory/
├── report.md # Markdown report with scoring and rationale
├── validation_report.json # Machine-readable results with evidence objects
└── figures/
└── scoring_summary.png # Bar chart of sub-scores with decision
When the user asks "Is [target] a good target for [disease]?":
target field and
an evidence block with at least one dimension populated.report.md, validation_report.json,
and figures/scoring_summary.png to the output directory.Demo mode (--demo): Uses pre-cached TGFBR1/IPF evidence — no API calls needed.
This is how judges and new users verify the skill works.
Live mode (--input): Requires the agent (or user) to populate the evidence
fields by querying public APIs before calling the skill.
These are the scientific rules encoded in this skill. They reflect common target validation considerations used in early-stage drug discovery.
| Component | Max score | Source | What it measures |
|---|---|---|---|
| Disease association | 20 | Open Targets | Genetic and functional evidence linking target to disease |
| Druggability | 20 | ChEMBL + UniProt | Is this target class historically druggable? Known ligands? |
| Chemical matter | 20 | ChEMBL | Do bioactive compounds exist? Best potency? |
| Clinical precedent | 20 | ChEMBL + ClinicalTrials.gov | Have compounds reached clinical trials? |
| Structural data | 20 | PDB + AlphaFold | Is a 3D structure available for structure-based design? |
If a target has strong disease relevance but also major systemic safety liability, prefer CONDITIONAL_GO over GO.
Safety penalties reduce the final score but do not change sub-scores. A target can score 80 on evidence but drop to 65 after safety adjustment. Safety is treated as a post-hoc penalty rather than a scoring dimension to ensure that strong biological evidence is not masked by safety concerns, but explicitly adjusted.
| Adjusted score | Decision | Meaning |
|---|---|---|
| 75-100 | GO | Strong evidence across multiple dimensions |
| 50-74 | CONDITIONAL_GO | Proceed with explicit risk mitigation plan |
| 25-49 | REVIEW | Insufficient evidence; needs more data |
| 0-24 | NO_GO | Target lacks fundamental validation |
Thresholds are calibrated to reflect typical target progression stages in early drug discovery, where strong multi-dimensional evidence (>=75) is required for full commitment.
Every piece of evidence is tagged with a confidence tier:
Evidence tiers guide confidence weighting and highlight where decisions rely on weaker or indirect evidence, enabling domain experts to focus review effort.
| Tier | Meaning | Example |
|---|---|---|
| T1 | Experimentally validated | Clinical trial data, GWAS with p < 5e-8 |
| T2 | Computational + literature supported | Known drug-target interaction with published SAR |
| T3 | Computationally predicted only | Docking score, ML prediction |
| T4 | Inferred or indirect | Pathway membership, guilt-by-association |
null with confidence: low, not scored as 0.The agent (LLM) dispatches and explains. The skill (Python) executes. The agent must NOT override scoring thresholds, invent gene-drug associations, skip safety warnings, or claim that a NO_GO target is worth pursuing. The skill does not replace wet-lab validation, medicinal chemistry review, or clinical judgement.