drug-design

End-to-end drug discovery pipeline orchestration. Deterministic Python script that auto-chains structure prediction, pocket detection, de novo design, docking, scoring, and ADMET filtering into reproducible workflows.

Quality

78%

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/drug-design/SKILL.md

Quality

Content

96%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 an exemplary operational reference: fully executable commands for every mode, explicit stage sequencing with documented validation and error-recovery loops, and no wasted tokens. The only structural improvement is offloading the per-mode stage and I/O schema detail into a reference file to slim the always-loaded SKILL.md.

Suggestions

Move the five mode-specific stage tables and the I/O contract JSON schemas into a references/pipeline-stages.md file, keeping a summary table and mode-selection guidance inline in SKILL.md.

Add brief instructions for inspecting and verifying _script_manifest.jsonl and pipeline_report.json after a run, so the critique/verification workflow is actionable from the skill itself.

Dimension	Reasoning	Score
Conciseness	The body is lean and entirely operational — tables, commands, and schemas with zero explanation of known concepts; even the 'Why a script instead of manual chaining' section conveys novel design rationale rather than padding. Every token earns its place.	5 / 5
Actionability	Copy-paste-ready commands cover all five modes plus sequence-only invocation, a complete argument table, concrete JSON schemas with the critical field flagged ("pockets[0].center must be [float, float, float]"), and executable error fixes like 'pip install vina' and the RDKit SMILES check.	5 / 5
Workflow Clarity	Per-mode stage tables give explicit input/output sequencing, the orchestrator's built-in validation is documented ("Validates I/O contracts between stages", "Stops on first failure with clear diagnostics"), and an Error Recovery table provides error → cause → fix feedback loops.	5 / 5
Progressive Disclosure	The bundle is well structured (single real script, scripts/pipeline.py, correctly referenced and present) with clearly signaled sections. However, ~100 lines of per-mode stage tables and JSON I/O schemas are inlined that could live in one-level-deep reference files, matching 'good structure; minor organization gaps' rather than the fully split top anchor.	4 / 5
	Total	19 / 20 Passed

Description

60%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 comprehensive about what the skill does, listing six concrete pipeline stages in third person. Its main weakness is trigger guidance: no 'Use when' clause and few natural trigger phrases, which both limits discoverability and caps completeness.

Suggestions

Add an explicit 'Use when...' clause with natural trigger phrases, e.g. 'Use when the user wants to find drugs, screen a compound library, run a docking campaign, or assess target druggability.'

Include common user synonyms and file extensions in the description (virtual screening, lead optimization, .pdb, .sdf) to improve trigger-term coverage.

Sharpen distinctiveness by contrasting with the standalone skills, e.g. 'orchestrates the full multi-stage workflow (use molecular-docking alone for single docking jobs)'.

Dimension	Reasoning	Score
Specificity	The description lists six concrete chained actions — "structure prediction, pocket detection, de novo design, docking, scoring, and ADMET filtering" — giving comprehensive, specific coverage with no gaps, matching the top anchor.	5 / 5
Completeness	The 'what' is clear (orchestrates the pipeline and names each stage), but there is no 'Use when...' clause or equivalent trigger guidance, so completeness is capped at 3 per the judging guidelines.	3 / 5
Trigger Term Quality	It contains relevant domain keywords ("drug discovery", "docking", "de novo design", "ADMET") but omits natural user phrases like "find drugs", "screen compounds", "virtual screening", "lead optimization", and file extensions (.sdf, .pdb), fitting the 'some relevant keywords but missing common variations' anchor.	3 / 5
Distinctiveness Conflict Risk	"End-to-end drug discovery pipeline orchestration" frames a niche, but terms like docking, scoring, and de novo design overlap significantly with the standalone related skills it chains (molecular-docking, binding-affinity, denovo-design), leaving real overlap risk for single-stage requests.	3 / 5
	Total	14 / 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/drug-design/SKILL.md
Commit: 4082a2e

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.