diffdock

DiffDock and DiffDock-L molecular docking. Use for protein-small-molecule pose prediction from PDB or sequence plus SMILES/SDF/MOL2, batch docking, virtual screening, and pose-confidence interpretation. Not for binding affinity prediction.

Quality

89%

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

Quality

Content

78%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 and well-structured with verified progressive-disclosure references, but carries noticeable verbosity (promotional overview, duplicated capability list, generic best-practices padding). Workflow sequencing is strong with validation present, though the batch feedback loop is only implicit.

Suggestions

Trim the Overview's promotional language ('state-of-the-art', 'crucial for structure-based drug discovery') and remove the 'Core Capabilities' list that duplicates the frontmatter description to reduce token cost.

Collapse the duplicated `temp_sampling_tor: 7.04` bullets into one line showing both the increase (flexible) and decrease (rigid) guidance.

Add an explicit validate->fix->retry feedback loop to the batch workflow steps (e.g., 'If --validate reports errors, fix the CSV and re-run --validate before docking') to lift workflow_clarity to 5.

Dimension	Reasoning	Score
Conciseness	Mostly actionable but padded: the Overview uses promotional language ('state-of-the-art', 'crucial for'), 'Core Capabilities' restates the frontmatter description, the 10-item Best Practices list is generic advice, and the temp_sampling_tor default is duplicated across two bullets.	3 / 5
Actionability	Provides copy-paste-ready `python -m inference` commands with real flags, a concrete CSV format, analysis-script invocations, and working ensemble-docking Python — fully executable and covering the common single, batch, screening, and analysis cases.	5 / 5
Workflow Clarity	Batch workflow is clearly sequenced (Step 1 prepare/validate CSV, Step 2 run) with validation checkpoints (setup_check.py, --validate), so it escapes the missing-validation cap; however the validate->fix->retry feedback loop is not shown inline in the batch steps.	4 / 5
Progressive Disclosure	SKILL.md is a clear overview with well-signaled one-level-deep references ('Read references/...') and a Resources section giving 'Read this file when users need...' guidance; all eight referenced bundle files (references, scripts, assets) are present and real.	5 / 5
	Total	17 / 20 Passed

Description

100%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 concise yet comprehensive, clearly stating what DiffDock does, when to use it, the supported input formats, and an explicit out-of-scope boundary. It hits every anchor at the top level with no padding.

Dimension	Reasoning	Score
Specificity	Names multiple concrete actions — 'pose prediction from PDB or sequence plus SMILES/SDF/MOL2', 'batch docking', 'virtual screening', 'pose-confidence interpretation' — giving comprehensive coverage of the skill's capabilities.	5 / 5
Completeness	Explicitly answers both 'what' (docking, pose prediction, batch, screening, confidence interpretation) and 'when' ('Use for...') with concrete trigger phrases, and adds a 'Not for binding affinity prediction' boundary.	5 / 5
Trigger Term Quality	Covers natural user phrasing ('molecular docking', 'pose prediction', 'virtual screening') plus synonyms and file extensions (PDB, SMILES/SDF/MOL2), matching the comprehensive-coverage anchor.	5 / 5
Distinctiveness Conflict Risk	Clear niche (DiffDock molecular docking) with distinct triggers; the explicit 'Not for binding affinity prediction' boundary minimizes conflict with affinity-scoring skills.	5 / 5
	Total	20 / 20 Passed

Validation

100%

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 — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: K-Dense-AI/scientific-agent-skills
Path: skills/diffdock/SKILL.md
Commit: 9cf7d9a

Reviewed: 2 days 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.