CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

pubchem-database

Query PubChem, search by name/CID/SMILES, retrieve properties, similarity/substructure searches, bioactivity, for cheminformatics. Use when a user asks about a specific chemical, drug, or molecule.

Quality

85%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

SKILL.md
Quality
Evals
Security

Quality

Content

82%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.

A well-structured, highly actionable skill body: all twelve capabilities are demonstrated with concrete executable commands, detailed material is correctly pushed to one-level-deep reference files, and fallback strategies handle failure cases. The main gaps are minor — light marketing phrasing ('robust', 'safe multi-agent use') that could be trimmed, no explicit result-validation checkpoints, and an orphaned citation.bib bundle file.

DimensionReasoningScore

Conciseness

The body is largely efficient — each of the 12 capabilities is one sentence plus one executable command with no explanation of concepts Claude already knows. Minor instances of over-explanation could be trimmed, e.g. "a robust Python wrapper script" and "This script allows safe multi-agent use of APIs", which is why it fits anchor 4 rather than anchor 5's 'every token earns its place'.

4 / 5

Actionability

Every capability comes with a copy-paste-ready command using real argument values (e.g. `uv run scripts/pubchem_api.py properties --cid 2244 --output result.json`, `resolve --inchi "InChI=1S/C3/c1-3-2/i1+1"`), and the specific examples cover the common cases for each subcommand. This matches anchor 5; it is above anchor 4 because there are no gaps — argument names, example values, and output handling are all concrete.

5 / 5

Workflow Clarity

Read-only API queries are organized into clear sections (resolution -> properties -> advanced searches), the Fallback Search Strategies section provides explicit error-recovery paths ("If searching for an ion or salt, try searching for the neutral parent compound"), and multi-step processes are delegated to the checklists in references/workflows.md. Not anchor 5 because the in-body guidance lacks explicit validation checkpoints on query results (e.g., checking for empty/error JSON before synthesizing an answer), though the workflows.md checklists partially compensate; not anchor 3 because sequence and recovery paths are clearly present.

4 / 5

Progressive Disclosure

The body is a well-organized overview with clearly signaled one-level-deep references — [references/endpoints.md](references/endpoints.md) for raw API construction and [references/workflows.md](references/workflows.md) for multi-step checklists, both verified to exist with no nested references. It falls short of anchor 5 only because references/citation.bib exists in the bundle but is never referenced or navigable from the body, a minor organization gap consistent with anchor 4.

4 / 5

Total

17

/

20

Passed

Description

83%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.

A strong description: it names the specific database and concrete capabilities, uses third person, and includes an explicit 'Use when' trigger clause with natural keywords. The only weakness is that the trigger clause is narrower than the capability list, leaving structure-search and property-query scenarios without explicit trigger coverage.

DimensionReasoningScore

Specificity

The description lists multiple concrete actions — "Query PubChem, search by name/CID/SMILES, retrieve properties, similarity/substructure searches, bioactivity" — which comprehensively covers the skill's capabilities with concrete search modalities and output types. It matches the anchor for multiple specific concrete actions with comprehensive coverage; score 4 would require minor gaps in coverage, but the capability list spans resolution, properties, and structure/bioactivity searches.

5 / 5

Completeness

Both parts are present: a clear 'what' ("Query PubChem, search by name/CID/SMILES, retrieve properties, similarity/substructure searches, bioactivity") and an explicit 'when' ("Use when a user asks about a specific chemical, drug, or molecule"). Not anchor 5 because the trigger clause covers only the lookup use case — users initiating similarity/substructure searches or property-range queries with a SMILES they already have are not explicitly covered by the 'when'.

4 / 5

Trigger Term Quality

Includes strong natural terms users would say — "chemical", "drug", "molecule" — plus domain-specific identifiers (name/CID/SMILES) and the "PubChem" brand itself. It falls short of anchor 5 because common synonyms like "compound", "substance", "medication", or "molecular structure" are missing, but coverage is well above anchor 3's 'some relevant keywords'.

4 / 5

Distinctiveness Conflict Risk

"Query PubChem" plus the cheminformatics domain (CID/SMILES, similarity/substructure, bioactivity) establishes a clear niche with distinct triggers (chemical, drug, molecule) and minimal overlap with other skills. It clearly matches the anchor for a clear niche with distinct triggers; nothing pushes it toward anchor 4's 'minor overlap risk with closely related skills'.

5 / 5

Total

18

/

20

Passed

Validation

93%

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

Validation — 15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

metadata_version

'metadata.version' is missing

Warning

Total

15

/

16

Passed

Repository

google-deepmind/science-skills

Path

skills/pubchem_database/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.