CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

molecular-rag

Retrieve structurally similar compounds with known properties from ChEMBL/ZINC to ground predictions and inform optimization. Based on MolRAG (Xian 2025, ACL).

64

Quality

80%

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 ./backend/cli/skills/chemistry/molecular-rag/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

93%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 lean skill document: executable commands verified against the actual script, clear scope boundaries, and a clean one-level-deep pointer to its bundle file. The only gap is the absence of guidance on interpreting outputs or handling empty results, which keeps workflow clarity just below top marks.

DimensionReasoningScore

Conciseness

The body is lean and assumes competence: a two-sentence overview, terse when-to-use/do-not-use bullets, an install command, three command examples, and a one-row script table. There is no re-teaching of concepts Claude already knows and no padding, matching the lean/every-token-earns-its-place anchor rather than the minor-trimmable-excess of 4.

5 / 5

Actionability

All three workflows are complete, copy-paste-ready bash commands, and every documented flag (--smiles, --similarity-threshold, --max-results, --target, --include-activities, --output) exists in the actual scripts/retrieve_analogs.py. The install command and script reference table round out fully executable guidance, matching anchor 5.

5 / 5

Workflow Clarity

The three workflows are unambiguous single commands with clear use/do-not-use boundaries, but there are no checkpoints or guidance for what to do with the output — e.g., interpreting analogs.json, or handling a query that returns no analogs. That fits anchor 4 (clear with minor validation gaps) rather than 5 (explicit checkpoints/feedback loops); the read-only nature means the destructive/batch cap does not apply.

4 / 5

Progressive Disclosure

The body is short, cleanly sectioned, and points one level deep to a real bundle file (scripts/retrieve_analogs.py) that is clearly signaled via the Script Reference table. There is no inlined bulk content or nested referencing, matching the well-signaled one-level-deep structure of anchor 5.

5 / 5

Total

19

/

20

Passed

Description

61%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 states a clear, domain-specific capability but omits any explicit "when to use" trigger guidance, capping completeness. Trigger keywords are good though missing common synonyms like "analogs" and "SAR". Distinctiveness is solid with minor overlap risk against sibling database skills.

Suggestions

Add an explicit trigger clause, e.g. "Use when predicting molecular properties, checking novelty of generated compounds, or grounding SAR reasoning in experimental data."

Include natural user phrasings such as "analogs", "analog search", "SAR", and "lead optimization" in the description to improve trigger term coverage.

Trim the trailing provenance sentence "Based on MolRAG (Xian 2025, ACL)" from the description (it is already in the body) so the description stays focused on capabilities and triggers.

DimensionReasoningScore

Specificity

"Retrieve structurally similar compounds with known properties from ChEMBL/ZINC" names the domain plus one concrete action, while "to ground predictions and inform optimization" are purposes rather than additional capabilities. This matches the anchor for 1-2 concrete actions without comprehensive coverage, not the several-specific-actions level of 4.

3 / 5

Completeness

The "what" is clear (retrieve similar compounds with known properties from ChEMBL/ZINC), but there is no "Use when..." clause or equivalent explicit trigger guidance; "when" is only weakly implied by the purpose phrasing. Per the judging guideline, a missing 'Use when' clause caps completeness at 3.

3 / 5

Trigger Term Quality

Natural terms like "structurally similar compounds", "ChEMBL", "ZINC", "predictions", and "optimization" give good coverage a user would plausibly say, exceeding the sparse single-keyword anchor of 3. However, common synonyms such as "analogs", "SAR", or "property prediction" are absent, so it falls short of the comprehensive synonym coverage of 5.

4 / 5

Distinctiveness Conflict Risk

The similarity/analog-retrieval framing carves a mostly distinct niche, but retrieval "from ChEMBL/ZINC" risks minor overlap with closely related database skills such as chembl-database. It is clearly more distinct than the broad overlap of anchor 3 but not the minimal-conflict clarity of anchor 5.

4 / 5

Total

14

/

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.

Validation — 14 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

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/molecular-rag/SKILL.md

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.