CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

clinical-trials-database

Query ClinicalTrials.gov via APIv2. Use when you want to search for trials by condition, drug, location, status, or phase; retrieve trial details by NCT ID; check eligibility/inclusion criteria; count trials across conditions or time periods; identify a sponsor's trial portfolio; find recruiting trials for patient matching.

Quality

90%

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 strong, highly actionable skill body: every command is executable with concrete worked examples, and detail is appropriately pushed to real, one-level-deep reference files. The main improvements are removing the duplicated --fields guidance and inline flag/enum reference (both covered by the reference files) and adding pagination-termination and empty-result guidance.

Suggestions

Collapse the 'Context Efficiency Warning' section into the existing Core Rules bullet on --fields to remove the duplicated advice, and trim the inline flag/enum reference to the most-used flags since references/clinical_trials_api.md already documents the full parameter and enum reference.

Add explicit pagination-termination guidance (e.g., 'stop when the response contains no nextPageToken') and a zero-result checkpoint (e.g., 'if totalCount is 0, broaden or remove filters') to strengthen workflow validation.

Either reference references/citation.bib from the body (e.g., when citing trial data in reports) or remove it from the bundle so every bundled file is discoverable.

DimensionReasoningScore

Conciseness

The body is largely operational (commands, flags, worked examples) and the Essie explanation is domain-specific and justified, but the '--fields' advice is repeated across 'Core Rules', the 'Context Efficiency Warning' section, the flag reference, and a TIP, and the full flag/enum reference duplicates what references/clinical_trials_api.md already documents. Fits anchor 4 ('minor instances of over-explanation that could be trimmed') rather than 5 due to this duplication, and not 3 since nearly all content is executable guidance.

4 / 5

Actionability

Every section provides copy-paste-ready 'uv run scripts/clinical_trials_api.py ...' commands, and specific worked examples cover the common cases (recruiting Phase 3 pediatric cystic fibrosis trials, atezolizumab for esophageal cancer, get-eligibility with a real NCT ID, pagination with an actual page token, Essie expressions). Fully matches the anchor-5 'copy-paste ready commands; specific examples cover the common cases'.

5 / 5

Workflow Clarity

A clear sequence exists ('Use --count-total first — check result volume before fetching all records', then '--fields', then pagination with '--page-token'), and commands are organized by use case with 'Use for:' preambles. It falls short of anchor 5 because there are no explicit validation/termination checkpoints (e.g., when nextPageToken is absent, or what to do when totalCount is 0), though the read-only nature of queries makes the destructive/batch cap inapplicable.

4 / 5

Progressive Disclosure

Referenced files (references/clinical_trials_api.md, references/studies_schema.md) exist, are exactly one level deep with no nested references, and are well signaled via '> [!TIP]' callouts and a closing References section. Minor gaps keep it at anchor 4 rather than 5: the ~35-line inline flag reference overlaps the parameter reference file, and references/citation.bib is present in the bundle but never referenced in the body.

4 / 5

Total

17

/

20

Passed

Description

95%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.

An excellent description: concrete, comprehensive, with an explicit Use-when trigger clause covering the main use cases and a clearly distinct niche. The only flaw is the second-person phrasing ('Use when you want to...'), which costs it the top specificity score under the rubric's voice rule.

DimensionReasoningScore

Specificity

The description lists six concrete actions ('search for trials by condition, drug, location, status, or phase; retrieve trial details by NCT ID; check eligibility/inclusion criteria; count trials...; identify a sponsor's trial portfolio; find recruiting trials for patient matching'), matching the comprehensive anchor-5 example. However, 'Use when you want to search' is second person voice, which the rubric penalizes by reducing specificity by 1, yielding 4.

4 / 5

Completeness

Both 'what' ('Query ClinicalTrials.gov via APIv2') and 'when' (an explicit 'Use when you want to...' clause enumerating six concrete trigger scenarios) are clearly and explicitly answered, matching the anchor-5 example's structure. Not 4, because the 'when' clause is fully explicit with concrete trigger phrases rather than merely adequate.

5 / 5

Trigger Term Quality

Natural user phrasing is comprehensively covered: 'trials', 'condition', 'drug', 'phase', 'status', 'eligibility/inclusion criteria' (explicit synonym pair), 'NCT ID', 'sponsor', 'recruiting', and 'patient matching' are all terms users would naturally say when needing this skill. Fits the comprehensive-synonyms anchor rather than anchor 4 because no commonly used domain term is missing.

5 / 5

Distinctiveness Conflict Risk

Clear niche (ClinicalTrials.gov clinical trial data) with distinct triggers such as 'NCT ID', 'sponsor's trial portfolio', and 'eligibility/inclusion criteria' that no generic skill would claim, giving minimal conflict risk. Fits the anchor-5 'clear niche with distinct triggers' example.

5 / 5

Total

19

/

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/clinical_trials_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.