CtrlK
BlogDocsLog inGet started
Tessl Logo

boltz-small-molecule-adme

Predict Tier-1 ADME/ADMET for small molecules with Boltz from bare SMILES — no target, no docking. Use when the user wants solubility, permeability, or lipophilicity/logD for a molecule or list of molecules. Not for ranking molecules against a protein target (use boltz-small-molecule-screen, which already returns ADME free).

75

Quality

92%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Critical

Do not install without reviewing

SKILL.md
Quality
Evals
Security

Quality

Content

85%

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

A highly actionable, well-structured skill with a clear validated workflow and clean progressive disclosure via real reference files. The only weakness is internal redundancy across sections that inflates token usage without adding information.

Suggestions

Dedupe the cost guidance: state the $0.01/molecule rate and "estimate-cost returns the authoritative total" once (e.g. in the cost step) and reference it from "Always Do This" instead of repeating both phrases three times.

State the 128-molecule batch cap and chunk/merge procedure once in the workflow and link to references/api.md rather than restating it in step 2, the command comments, and two "Always Do This" bullets.

Consolidate the Claude Code vs Codex synchronous/polling note into a single location; the near-verbatim paragraph appears in both the command-pattern comment and the "Always Do This" list.

DimensionReasoningScore

Conciseness

The body is lean and avoids explaining concepts Claude already knows, but the same guidance is restated across sections — the 128-molecule cap appears ~4 times, the $0.01/molecule and "estimate-cost is authoritative" note ~3 times, and the Codex synchronous/polling instruction twice near-verbatim — so it could be tightened.

2 / 3

Actionability

Provides a fully executable CLI invocation with all required flags, concrete absolute-path placeholders, an example run name, and exact output field names — copy-paste ready.

3 / 3

Workflow Clarity

The five numbered steps form a clear sequence with explicit validation checkpoints (estimate-cost → wait for confirmation → run → check per-molecule status) and an error-recovery feedback loop for failed SMILES.

3 / 3

Progressive Disclosure

SKILL.md is a concise overview with well-signaled, one-level-deep references to references/api.md and references/results.md (both verified to exist), with payload and output-layout detail appropriately split out.

3 / 3

Total

11

/

12

Passed

Description

100%

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

A strong, third-person description that states concrete capabilities, explicit "Use when" triggers, and clear disambiguation from the related screening skill. It hits all four anchors at the top level.

DimensionReasoningScore

Specificity

Lists multiple concrete capabilities — "Predict Tier-1 ADME/ADMET for small molecules with Boltz from bare SMILES" with specific properties solubility, permeability, and lipophilicity/logD — matching the anchor for multiple specific concrete actions.

3 / 3

Completeness

Explicitly answers both what ("Predict Tier-1 ADME/ADMET...") and when ("Use when the user wants solubility, permeability, or lipophilicity/logD..."), matching the anchor for clear what-and-when with explicit triggers.

3 / 3

Trigger Term Quality

Includes natural user-facing terms ("solubility, permeability, or lipophilicity/logD for a molecule or list of molecules") alongside ADME/ADMET and SMILES, giving good coverage of phrasing a user would actually say.

3 / 3

Distinctiveness Conflict Risk

Actively disambiguates with "Not for ranking molecules against a protein target (use boltz-small-molecule-screen, which already returns ADME free)", carving a clear niche unlikely to trigger the wrong skill.

3 / 3

Total

12

/

12

Passed

Validation

100%

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

Validation16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository
openai/plugins
Reviewed

Table of Contents

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.