CtrlK
BlogDocsLog inGet started
Tessl Logo

adme-property-predictor

Analyze data with `adme-property-predictor` using a reproducible workflow, explicit validation, and structured outputs for review-ready interpretation.

35

Quality

31%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./scientific-skills/Data Analysis/adme-property-predictor/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

38%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 content is a long, partly-templated document that mixes genuine CLI guidance with fabricated Python API examples and inlined reference material, while pointing to reference and script files that do not exist in the bundle. It is structured but not reliably executable or navigable.

Suggestions

Reconcile all code examples with the actual bundle: remove the non-existent ADMEPredictor/BatchProcessor API and the --filter/--rank-by/--top-n/--format csv flags, or add the corresponding modules and argparse options to main.py.

Move the property tables, best-practices, troubleshooting, performance, and limitations blocks into the referenced files (qsar_models.md, property_ranges.md, etc.) and actually create those files so the References links resolve.

Trim the templated boilerplate at the top (When to Use bullet 1, Key Features, Implementation Details) so the body leads with the real CLI usage and a single concrete workflow.

DimensionReasoningScore

Conciseness

The ~700-line body is noticeably verbose: the top ~60 lines are templated boilerplate ("Scope-focused workflow aligned to: [description]", "See `## Workflow` above", a garbled "When to Use" bullet) and large reference-style blocks (property tables, troubleshooting, performance, model-accuracy stats) are inlined rather than deferred.

2 / 5

Actionability

There is concrete guidance and the real `main.py` CLI is documented (py_compile, --help, --smiles/--properties/--format/--input/--output), but most examples are not executable: the Python API (ADMEPredictor.predict_absorption, BatchProcessor, batch_predict/filter/rank) and CLI flags (--filter, --rank-by, --top-n, --format csv) reference modules and options that do not exist in the bundled script.

3 / 5

Workflow Clarity

A sequenced Workflow with validation (scope check, fallback path) and a Quality Checklist exist, but the workflow is fragmented across four overlapping sections (Workflow, Example run plan, Implementation Details, Complete Workflow Example) and the concrete execution path uses fabricated commands, leaving the actual steps muddy.

3 / 5

Progressive Disclosure

Although the body has many section headers, hundreds of lines of API/reference material that belong in separate files are inlined, and the navigation contract is broken: the References section lists 6 files and Scripts lists 8 files, but only references/runtime_checklist.md and scripts/main.py actually exist.

2 / 5

Total

10

/

20

Passed

Description

25%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 reads as generic process boilerplate rather than a statement of the skill's actual pharmacokinetic capability, omitting the ADME/drug-likeness niche entirely. It also lacks any explicit "Use when" trigger guidance, which both limits discoverability and raises conflict risk.

Suggestions

State the concrete capability: e.g. "Predict ADME (absorption, distribution, metabolism, excretion) and drug-likeness properties of small molecules from SMILES strings."

Add an explicit "Use when" clause with natural triggers: "Use when the user asks about ADME properties, pharmacokinetics, drug-likeness, Lipinski/Veber rules, or batch screening of compound libraries (SMILES/.smi/.sdf)."

Drop the generic process language ("reproducible workflow", "structured outputs") from the description; it does not distinguish this skill from any other data-analysis skill.

DimensionReasoningScore

Specificity

The description names the domain ("Analyze data") but the actions are generic process descriptors ("reproducible workflow", "explicit validation", "structured outputs") rather than concrete capabilities, and it never states what the skill actually predicts (ADME/drug-likeness properties).

2 / 5

Completeness

It offers a vague "what" ("Analyze data ... using a reproducible workflow") and no "when" at all — there is no "Use when..." clause or equivalent trigger guidance, matching the anchor for a vague what with absent when.

2 / 5

Trigger Term Quality

Only generic keywords ("Analyze data", "validation", "structured outputs") appear; the natural terms a user would say (ADME, pharmacokinetics, drug-likeness, SMILES, molecules) are absent, and `adme-property-predictor` is the skill name rather than a user-spoken trigger.

2 / 5

Distinctiveness Conflict Risk

The descriptive text is very broad ("Analyze data ... reproducible workflow, explicit validation, structured outputs") and would overlap with most data-analysis skills; only the embedded skill name provides any niche signal.

2 / 5

Total

8

/

20

Passed

Validation

87%

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

Validation14 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

skill_md_line_count

SKILL.md is long (714 lines); consider splitting into references/ and linking

Warning

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

14

/

16

Passed

Repository
aipoch/medical-research-skills
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.