admet-prediction

ADMET property prediction for drug candidates. Full pharmacokinetic panel (Caco-2, PPB, clearance, CYP), toxicity (hERG, AMES, DILI), drug-likeness (Lipinski, QED), using RDKit descriptors and TDC models.

Quality

73%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./backend/cli/skills/chemistry/admet-prediction/SKILL.md

Quality

Content

72%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is highly actionable with verified, copy-paste-ready commands and good organization, and is mostly token-efficient. It loses points for omitting validation checkpoints on batch CSV operations and for failing to link the existing endpoints reference file.

Suggestions

Add an explicit validation/verification step to the batch workflows (e.g., 'Confirm the CSV parsed N molecules before predicting; spot-check RED flags in the output'), since batch operations are capped at 3 without validation.

Link `references/endpoints_guide.md` from the body (e.g., under an 'Endpoint reference' section: 'See [references/endpoints_guide.md] for per-endpoint clinical significance and thresholds') so the one-level-deep reference is clearly signaled.

Trim the Overview's background on what ADMET is and the 40%-of-failures statistic, since Claude already knows this; keep the overview focused on what the skill does.

Dimension	Reasoning	Score
Conciseness	Largely efficient and well-structured with tables and terse command blocks; the Overview does briefly re-explain ADMET and a clinical-failure statistic that Claude already knows, which is minor over-explanation that could be trimmed.	4 / 5
Actionability	Every workflow ships copy-paste-ready, executable CLI commands (e.g., `python scripts/predict_admet.py --input ... --format csv`) verified against the scripts' argparse, plus a script-reference table and a concrete threshold table covering the common cases.	5 / 5
Workflow Clarity	Workflows are sequenced into numbered sections, but they drive batch CSV-in/CSV-out operations with no documented validation or verification step (e.g., checking input molecules parsed, confirming output written, sanity-checking RED flags); per the rubric, missing validation for batch operations caps this dimension at 3.	3 / 5
Progressive Disclosure	The body is well-sectioned, but the bundled `references/endpoints_guide.md` is never referenced or linked from SKILL.md, so a one-level-deep reference is present but not signaled; a substantial reference file that should be navigable is effectively buried.	3 / 5
	Total	15 / 20 Passed

Description

75%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is specific and well-differentiated, with comprehensive coverage of ADMET capabilities and strong domain keywords. Its main weakness is the absence of an explicit 'Use when...' trigger clause, which caps completeness at 3.

Suggestions

Add an explicit 'Use when...' clause (e.g., 'Use when predicting pharmacokinetic or toxicity properties of drug candidates, or when ranking compounds by ADMET profile') to lift completeness above 3.

Include a few natural-language trigger phrases or synonyms (e.g., 'drug-likeness screening', 'compound profiling') alongside the technical endpoint names to broaden trigger-term quality.

Optionally mention supported input formats (SMILES, CSV) so users know when to reach for this skill versus a general chemistry tool.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete, specific actions covering the full ADMET scope: "Full pharmacokinetic panel (Caco-2, PPB, clearance, CYP), toxicity (hERG, AMES, DILI), drug-likeness (Lipinski, QED), using RDKit descriptors and TDC models", matching the comprehensive-coverage anchor.	5 / 5
Completeness	The 'what' is clearly and concretely stated, but there is no 'Use when...' clause or equivalent explicit trigger guidance, so completeness is capped at 3 per the rubric guideline.	3 / 5
Trigger Term Quality	Strong domain keyword coverage (ADMET, Caco-2, PPB, hERG, AMES, DILI, Lipinski, QED, RDKit, TDC), but it leans on technical jargon and omits natural synonyms or file extensions a user might say; a few natural trigger phrases are missing.	4 / 5
Distinctiveness Conflict Risk	Occupies a clear niche (ADMET property prediction for drug candidates) with distinct domain-specific triggers and minimal overlap risk with other skills.	5 / 5
	Total	17 / 20 Passed

Validation

87%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
metadata_version	'metadata.version' is missing	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	14 / 16 Passed

Repository: synthetic-sciences/openscience
Path: backend/cli/skills/chemistry/admet-prediction/SKILL.md
Commit: 3a6c3a9

Reviewed: 1 day ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.