CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

hmdb-skill

Submit compact HMDB search requests for metabolites, proteins, diseases, and pathways. Use when a user wants concise HMDB summaries

68

Quality

86%

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

92%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 tight, highly actionable skill body: executable examples, a precise input/output contract verified against the actual script, and clean section structure with no filler. The only defect is minor duplication — the same example request and base_url repeated — which costs a little token efficiency.

DimensionReasoningScore

Conciseness

The body is lean — no explanations of concepts Claude already knows, just operating rules, exact field lists, and concrete patterns. Not level 5 because of verifiable redundancy: the identical serotonin JSON example appears both under 'Common HMDB patterns' and again in the 'Execution' bash block, and 'base_url=https://hmdb.ca' is stated three times. Not level 3 or below because everything else earns its place.

4 / 5

Actionability

Fully executable guidance: a copy-paste-ready bash command with a complete JSON payload, two concrete request patterns covering the common cases (metabolites and pathways), an exhaustive required/optional field list, and explicit success/failure output shapes. The documented fields were verified against the actual script interface (base_url, path, record_path, max_items, save_raw, raw_output_path all exist in scripts/rest_request.py). It matches the 'fully executable; copy-paste ready; specific examples cover the common cases' anchor.

5 / 5

Workflow Clarity

This is a simple, single-purpose skill (pipe one JSON object to scripts/rest_request.py, return a concise summary) and the single action is unambiguous, so the simple-skill exception applies. The read-only search operations are neither destructive nor batch, so no validation cap applies; failure detection is even documented ('Failure returns ok=false with error.code and error.message').

5 / 5

Progressive Disclosure

The body is under 50 lines, well-organized into six clearly headed sections, needs no external references, and explicitly states 'No additional runtime references are required; keep the import package limited to this file and scripts/rest_request.py' — the single referenced bundle file exists. This matches the guideline that a short, reference-free skill with well-organized sections scores 5.

5 / 5

Total

19

/

20

Passed

Description

73%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 solid, third-person description with an explicit 'Use when' clause, concrete domain action, and low conflict risk. Its main weakness is a single-action 'what' and a narrow trigger clause that omits common user phrasings like 'look up a metabolite' or the full database name.

Suggestions

Broaden the trigger clause to cover more natural phrasings, e.g. 'Use when the user wants to look up or search HMDB for a metabolite, protein, disease, or pathway, or wants a concise summary of one'.

Add one more concrete capability to the 'what', such as 'and retrieve full raw payloads on demand', to move beyond a single action.

Consider including the full database name ('Human Metabolome Database') once as a synonym for HMDB to improve keyword coverage.

DimensionReasoningScore

Specificity

The description names the domain and one concrete action — 'Submit compact HMDB search requests for metabolites, proteins, diseases, and pathways' — with the four entity categories adding specificity, but it covers only a single action (submitting searches) rather than several distinct capabilities like retrieving raw payloads or formatting summaries. It matches the 'names domain and 1-2 concrete actions, but not comprehensive' anchor; it is not level 4 because no additional actions are listed, and not level 2 because the action is concrete and domain-scoped, not generic.

3 / 5

Completeness

Both parts are explicitly present: the 'what' ('Submit compact HMDB search requests for metabolites, proteins, diseases, and pathways') and the 'when' ('Use when a user wants concise HMDB summaries'). Not level 5 because the trigger clause is narrow and partially tautological — it only fires on wanting 'concise HMDB summaries' and misses natural triggers like looking up a specific metabolite or pathway. Not level 3 because an explicit 'Use when...' clause with a concrete trigger phrase exists.

4 / 5

Trigger Term Quality

Natural terms include 'HMDB search', 'metabolites', 'proteins', 'diseases', 'pathways', and 'concise HMDB summaries' — good coverage of the phrasing a user would naturally use. Not level 5 because common variations are missing: no 'lookup', 'look up', 'human metabolome database', or 'query' phrasings. Not level 3 because the keywords go beyond a bare domain mention and span all four content categories.

4 / 5

Distinctiveness Conflict Risk

'HMDB' is a clear, unique niche with distinct triggers covering four specific content categories; no other plausible skill would fire on these terms. It matches the 'clear niche with distinct triggers; minimal conflict risk' anchor exactly.

5 / 5

Total

16

/

20

Passed

Validation

100%

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

openai/plugins

Path

plugins/life-science-research/skills/hmdb-skill/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.