CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

bio-ontology-mapper

Map unstructured biomedical text to standardized ontologies (SNOMED CT.

Quality

54%

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

Fix and improve this skill with Tessl

tessl review fix ./scientific-skills/Evidence Insight/bio-ontology-mapper/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

38%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 delivers genuinely valuable domain content — capability walkthroughs, a quality checklist, pitfalls, limitations, and strong PHI/clinical-safety warnings — but wraps it in heavy duplicated template boilerplate and documents a bundle layout that does not match reality. The phantom references and non-executable Python examples are the most consequential defects, undermining both actionability and progressive disclosure.

Suggestions

Reconcile the 'References' and 'Scripts' sections with the actual bundle: document snomed_sample.json, mesh_sample.json, synonyms.json and main.py, or remove the nonexistent files; broken navigation is the top progressive-disclosure failure.

Replace the scripts.mapper / cross_map / batch_map examples with executable usage of the real main.py CLI (e.g. an actual 'python scripts/main.py --term "myocardial infarction"' invocation with sample output).

Delete the redundant 'Quick Check' and 'Audit-Ready Commands' sections (duplicates of Example Usage) and the boilerplate 'Key Features'/'Implementation Details' filler, and fill in the empty parameter descriptions ('--ontology', '--threshold', '--format', '--api-key').

DimensionReasoningScore

Conciseness

The body contains substantial template boilerplate: the description is repeated nearly verbatim in 'When to Use' and 'Key Features', the same 'python -m py_compile scripts/main.py' command appears in 'Example Usage', 'Quick Check', and 'Audit-Ready Commands', and filler lines like 'Structured execution path designed to keep outputs consistent and reviewable' add no information. This matches 'noticeably verbose; several unnecessary explanations or padded sections'; not a 1 because real, substantive content (capabilities, checklists, pitfalls, limitations) is present.

2 / 5

Actionability

There is a runnable CLI ('python scripts/main.py --help'), a parameters table, and specific safety guidance, but the flagship Python examples ('from scripts.mapper import BioOntologyMapper', 'mapper.cross_map', 'mapper.batch_map') reference modules that do not exist in the bundle — only scripts/main.py is present — so the code is not executable as written. Several parameter rows ('--ontology', '--threshold', '--format', '--api-key') also have empty descriptions. This matches 'some concrete guidance but incomplete ... missing key details'; not a 4 because the primary code examples cannot run against the actual bundle.

3 / 5

Workflow Clarity

A sequenced 5-step Workflow with a fallback path exists, plus pre/during/post-mapping checklists, but it is fragmented across three overlapping sections ('Example run plan', 'Workflow', 'Implementation Details') with a circular pointer ('See ## Workflow above'). Per the judging guidelines, the batch-processing capability has no validate-then-proceed feedback loop (py_compile is only a syntax check; there is no command-level validation of mapping output before returning it), capping workflow clarity at 3.

3 / 5

Progressive Disclosure

The 'References' section lists six markdown files (snomed_ct_guide.md, mesh_structure.md, etc.) and 'Scripts' lists eight modules (mapper.py, extractor.py, ...), none of which exist in the bundle; the actual references/ contains three JSON sample files (snomed_sample.json, mesh_sample.json, synonyms.json) that the body never mentions. Navigation points to phantom files, which is worse than 'references present but not clearly signaled' (3) and matches the minimal-structure anchor (2); it is not a 1 because the body does have sectioned organization and one-level-deep reference intent.

2 / 5

Total

10

/

20

Passed

Description

50%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 communicates a clear, domain-specific purpose but is visibly truncated ('...ontologies (SNOMED CT.'), omitting the rest of the ontology list and any 'when to use' trigger clause. It functions adequately for retrieval but falls short of the good examples' standard of complete capability lists plus explicit trigger conditions.

Suggestions

Complete the truncated sentence, e.g. 'Map unstructured biomedical text to standardized ontologies (SNOMED CT, MeSH, ICD-10, LOINC, RxNorm). Use when normalizing clinical terms, coding diagnoses, or translating concepts between medical terminologies.'

Add a 'Use when...' clause with natural trigger phrases such as 'normalize medical terms', 'SNOMED coding', or 'terminology crosswalk' to lift completeness and trigger-term quality.

Include a second concrete capability (e.g. confidence scoring or cross-ontology translation) to strengthen specificity beyond the single 'map' action.

DimensionReasoningScore

Specificity

The description names the domain ("unstructured biomedical text", "standardized ontologies") and one concrete action ("Map ... to standardized ontologies", naming SNOMED CT), matching the anchor for a domain plus 1-2 concrete actions. It is not a 4 because the sentence is truncated mid-list ('SNOMED CT.') so only a single action survives and coverage is incomplete; it is not a 2 because a real, concrete mapping action is stated rather than a generic verb.

3 / 5

Completeness

The 'what' is clear (map unstructured biomedical text to standardized ontologies) but there is no 'Use when...' clause or equivalent trigger guidance, which per the judging guidelines caps completeness at 3. It is not a 2 because the 'what' is concrete and specific rather than vague; it cannot be a 4 because 'when' is entirely absent and the sentence is cut off mid-parenthesis, leaving even the ontology list incomplete.

3 / 5

Trigger Term Quality

It includes some genuinely natural terms a user would say — "SNOMED CT", "biomedical text", "ontologies" — but misses common variations and synonyms such as "medical terms", "normalize", "terminology", "ICD-10", "MeSH", or "code lookup". This matches 'some relevant keywords but missing common variations or synonyms'; a 4 would require broader natural-term coverage, which the truncation prevents.

3 / 5

Distinctiveness Conflict Risk

Biomedical ontology mapping is a recognizable niche and 'SNOMED CT' is a fairly distinctive term, but the truncated description gives no explicit trigger phrases, so it could still overlap with other medical-terminology or data-harmonization skills. This sits between 'somewhat specific but could still overlap' (3) and 'mostly distinct' (4); the missing tail of the description and absence of distinct triggers keep it at 3.

3 / 5

Total

12

/

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

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

15

/

16

Passed

Repository

aipoch/medical-research-skills

Path

scientific-skills/Evidence Insight/bio-ontology-mapper/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.