CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

cosmic-database

Access COSMIC to download mutation datasets, query Cancer Gene Census, and retrieve mutational signatures when your genomic analysis requires curated somatic mutation resources.

Quality

70%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

High

Do not use without reviewing

Fix and improve this skill with Tessl

tessl review fix ./scientific-skills/Evidence Insight/cosmic-database/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

50%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 second half of the body contains genuinely useful, near-executable guidance for authenticated COSMIC downloads and pandas loading, with a real one-level-deep reference file. However, five entire boilerplate sections are duplicated at the top of the file, introducing contradictory dependency information and roughly doubling token cost, and the workflow lacks explicit validation checkpoints for the authenticated download step.

Suggestions

Delete the first five duplicated sections (the boilerplate When to Use / Key Features / Dependencies / Example Usage / Implementation Details block) and keep only the concrete second versions, removing the contradictory Python 3.10+ vs 3.9+ and dependency-list conflict in one stroke.

Add an explicit validation checkpoint to the typical workflow (e.g., verify the HTTP response succeeded and the downloaded file is non-empty/valid gzip before loading into pandas), especially since authentication failures are a common failure mode for this skill.

Fix the executable details: replace the environment-specific bash path ("20260316/scientific-skills/...") with a repo-relative one, and correct the import example so it runs as written (e.g., sys.path handling or invoking the script directly), since the script's own comments admit the API parameter structure is unverified.

DimensionReasoningScore

Conciseness

The body duplicates five entire sections (When to Use, Key Features, Dependencies, Example Usage, Implementation Details each appear twice, e.g. lines 11-17 vs 54-62 and 26-29 vs 72-81), with the first set being generic boilerplate ("Structured execution path designed to keep outputs consistent and reviewable") — noticeably verbose with several padded sections. Not 1 because it never explains background concepts Claude already knows and the second half is lean; not 3 because roughly half the file is redundant or filler.

2 / 5

Actionability

The second Example Usage offers near-executable Python (download_cosmic_file call with a concrete filepath plus pd.read_csv with compression="gzip"), but there are real execution gaps: `from scripts.download_cosmic import download_cosmic_file` assumes an import layout that does not work as written without package setup, the bash example hardcodes an environment-specific path ("20260316/scientific-skills/..."), and the script itself notes "the actual COSMIC API might use a different parameter structure" — so the documented happy path is not verified to run. This sits between concrete-with-gaps (4) and incomplete-guidance (3), landing at 3.

3 / 5

Workflow Clarity

"Typical workflow: 1. Download the desired COSMIC export. 2. Load into a DataFrame... 3. Filter/aggregate" provides a sequence, but validation checkpoints are missing or implicit (no step to verify authentication success or download completeness) and the duplicated sections give contradictory guidance (Python 3.10+ vs 3.9+; two different dependency lists). Not 4 because there is no explicit verify step and the contradiction muddies the sequence; not 2 because the sequence is specific and mostly coherent.

3 / 5

Progressive Disclosure

The bundle structure is sound — references/cosmic_data_reference.md and scripts/download_cosmic.py both exist, are one level deep, and the reference is clearly signaled (line 110) — but the five duplicated top-level section headers make the document confusing to navigate and split effectively the same content across two competing outlines. Structure present but organization is impaired rather than having only minor gaps.

3 / 5

Total

11

/

20

Passed

Description

78%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: third-person voice, three concrete actions, an explicit (if slightly circular) trigger clause, and a clearly distinct niche. Its main gaps are missing synonyms/format keywords (TSV, VCF, SBS/DBS/ID) and a when-clause that could enumerate concrete trigger scenarios.

DimensionReasoningScore

Specificity

"download mutation datasets, query Cancer Gene Census, and retrieve mutational signatures" lists several specific concrete actions, matching the anchor-4 level of coverage. Not 5 because it omits supported dataset types (copy number, fusions, expression) that the skill actually handles; not 3 because it clearly goes beyond 1-2 actions.

4 / 5

Completeness

Both "what" (the three download/query/retrieve actions) and "when" ("when your genomic analysis requires curated somatic mutation resources") are explicitly present. Not 5 because the trigger is a single conditional clause that borders on tautological rather than the concrete trigger phrases in the anchor-5 example; not 3 because the when-clause is explicit, not merely implied.

4 / 5

Trigger Term Quality

Natural domain terms a genomics user would say — "COSMIC", "mutation datasets", "Cancer Gene Census", "mutational signatures" — give good keyword coverage. Not 5 because common synonyms and format terms users mention (TSV/VCF, SBS/DBS/ID, cancer gene lists) are absent; not 3 because coverage is broad, not jargon-only.

4 / 5

Distinctiveness Conflict Risk

"COSMIC", "Cancer Gene Census", and "mutational signatures" name a clear niche with distinct triggers and minimal conflict risk with other skills — no other common skill would trigger on these phrases.

5 / 5

Total

17

/

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/cosmic-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.