CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

protein-sequence-msa

Performs multiple sequence alignment of proteins with EBI Clustal Omega. Use when you need to align multiple sequences to assess similarity, domain conservation, or key residue conservation. Supports up to 4000 sequences and a maximum file size of 4 MB. Do not use to search for homologous proteins in a database (use MMseqs2, BLAST), align non-protein sequences (DNA, RNA), perform structural alignment (use Foldseek, PyMOL), or if you only have a single sequence.

Quality

84%

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

75%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, actionable skill body with a real executable script, a concrete input example, and thoughtful domain guidance on metric selection. The main gaps are the missing output-format specification and parsing guidance for the interpretation step, the absence of any output-validation/error-recovery checkpoint, and an orphaned citation.bib that is never linked from the body.

Suggestions

Reference the bundle's references/citation.bib from the References section (e.g., when citing Clustal Omega in reports) so the file is discoverable.

Specify the output alignment file format and show a short snippet or command for computing identical matches / conserved columns from it, making the interpretation step fully executable.

Add a validation checkpoint after execution (e.g., verify the output file is non-empty and report EBI job errors) to close the workflow's feedback loop.

DimensionReasoningScore

Conciseness

The body is largely efficient — prerequisites, core rules, a copy-paste command, and a concrete FASTA example with no padding about what MSA is. Minor trimming opportunities: the "Goal" section restates the frontmatter description nearly verbatim, and some interpretation bullets ("Use these columns to drive the assessment") are wordier than needed. This fits the 4 anchor (minor over-explanation) rather than the lean 5 anchor.

4 / 5

Actionability

Mostly executable: the exact command `uv run scripts/msa_align.py <INPUT_FASTA> -o <OUTPUT_FILE>` (script present in the bundle), a concrete FASTA input example, and explicit metric formulas like `(Identical Residue Matches) / (Length of Shorter Sequence)`. It stops short of the 5 anchor because the output alignment format is never specified and no code or command shows how to compute identical matches / conserved columns from the alignment file.

4 / 5

Workflow Clarity

Clear sequence: prerequisites (uv, license notification, .env, credential) → prepare input with explicit format requirements → execute via wrapper → interpret and report, with numbered instructions and input constraints acting as pre-checks. It does not reach 5 because there is no post-run validation of the output or error-recovery loop (e.g., what to do if the EBI job fails or the output file is empty); this is not a destructive operation so the batch cap does not apply.

4 / 5

Progressive Disclosure

Good structure with clear sections (Prerequisites, Core Rules, Goal, Instructions, References), and the script path `scripts/msa_align.py` is well-signaled and exists in the bundle. Minor organization gap: `references/citation.bib` is present in the bundle but never referenced or linked from the body — the References section only lists the EBI URL — so it falls just short of the 5 anchor's well-signaled references.

4 / 5

Total

16

/

20

Passed

Description

87%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, explicit what/when, concrete capacity limits, and unusually good do-not-use disambiguation against neighboring tools. The only gaps are a few missing natural trigger terms ("MSA", "FASTA") and the fact that it describes one core action rather than a list of several.

DimensionReasoningScore

Specificity

The description states the concrete action ("Performs multiple sequence alignment of proteins with EBI Clustal Omega") plus concrete analytical purposes ("assess similarity, domain conservation, or key residue conservation") and explicit capacity limits ("up to 4000 sequences and a maximum file size of 4 MB"). It falls short of the 5 anchor because the core capability is a single action with purposes rather than multiple distinct concrete actions like the top anchor's extract/fill/merge/convert list.

4 / 5

Completeness

It explicitly answers both what ("Performs multiple sequence alignment of proteins with EBI Clustal Omega") and when ("Use when you need to align multiple sequences to assess similarity, domain conservation, or key residue conservation") with concrete trigger phrases, and goes further with explicit do-not-use guidance. This matches the 5 anchor exactly; the 4 anchor's "'when' could be more explicit" does not apply.

5 / 5

Trigger Term Quality

Good natural keyword coverage: "multiple sequence alignment", "align multiple sequences", "protein", "similarity", "domain conservation", "key residue conservation", "Clustal Omega". A few natural terms users would say are missing — the acronym "MSA" and the input format "FASTA" — so it does not reach the comprehensive-synonyms level of the 5 anchor.

4 / 5

Distinctiveness Conflict Risk

Clear niche with minimal conflict risk: the description explicitly disambiguates against adjacent tools ("Do not use to search for homologous proteins in a database (use MMseqs2, BLAST), align non-protein sequences (DNA, RNA), perform structural alignment (use Foldseek, PyMOL), or if you only have a single sequence"), which is stronger than the 4 anchor's minor-overlap profile.

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/protein_sequence_msa/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.