CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

pdb-database

Access the RCSB Protein Data Bank (PDB) to search, download, and programmatically retrieve 3D macromolecular structures and metadata; use when you need structure discovery (text/sequence/3D similarity) or automated structural data ingestion for structural biology and drug discovery workflows.

Quality

79%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

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

Quality

Content

68%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 is highly actionable with an excellent runnable end-to-end example and a sensible split to a single reference file. The main gaps are a redundant Key Features section and format explanations that pad token cost, plus missing validation checkpoints for the batch-retrieval workflow.

Suggestions

Remove or merge the "Key Features" section into "When to Use" to eliminate the duplicated bullet lists, and drop the PDB/mmCIF format primer in favor of a one-line format choice rule.

Add explicit validation to the batch pattern, e.g., check HTTP status / file size per ID, retry with backoff, and collect failed IDs for reporting — this would lift workflow_clarity above the batch cap of 3.

Replace the conditional "If present in this repository, consult" with a direct pointer like "See [references/api_reference.md](references/api_reference.md) for advanced query patterns, rate limits, and troubleshooting" to strengthen reference signaling.

DimensionReasoningScore

Conciseness

Mostly efficient, but the "Key Features" section ("Text and attribute-based search over RCSB PDB entries", "Sequence similarity search...", etc.) substantially duplicates the "When to Use" bullets, and lines like "PDB: legacy text format; widely supported but less expressive" / "mmCIF (PDBx): modern standard" explain concepts Claude already knows. This matches anchor 3 ("mostly efficient but includes some unnecessary explanation or could be tightened") better than 4, where over-explanation would be minor.

3 / 5

Actionability

The body provides a complete, copy-paste-ready end-to-end script (search with combined queries, metadata fetch, coordinate download, parsing), an install command, direct download URLs, and a working GraphQL example — fully executable guidance covering the common cases, matching anchor 5. Not 4 because no key details are missing; the examples run as written.

5 / 5

Workflow Clarity

The example script lays out a clear numbered sequence (search → fetch metadata → download → parse) and the batch section says to "handle exceptions per-ID to keep pipelines robust", but there are no explicit validation checkpoints for batch operations (e.g., verify each download, retry/fail-fast policy). The rubric caps workflow clarity at 3 for batch workflows lacking validation steps, so it cannot score 4 despite the good sequencing.

3 / 5

Progressive Disclosure

Good structure: overview content stays in SKILL.md and advanced material ("advanced endpoint usage, query patterns, schema notes, rate limits, and troubleshooting") is delegated to a real, verified one-level-deep reference (references/api_reference.md). It is not 5 because the reference pointer is hedged ("If present in this repository, consult"), which weakens the signaling, and some inline implementation detail (e.g., search-mode semantics) could live in the reference.

4 / 5

Total

15

/

20

Passed

Description

82%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, concrete description that clearly states capabilities and explicit use conditions with a clear niche. Its main weaknesses are second-person phrasing ("use when you need") and a few missing natural trigger synonyms such as file formats (PDB/mmCIF) or "protein structure".

Suggestions

Rewrite in third person to avoid the second-person trigger clause penalty, e.g., "...metadata; use for structure discovery (text/sequence/3D similarity)..." → "Use when a user needs structure discovery..." or drop 'you' entirely.

Add natural synonyms users would actually say, such as "protein structures", "PDB files", "mmCIF", or "homolog search", to broaden trigger coverage.

DimensionReasoningScore

Specificity

The description names concrete actions — "search, download, and programmatically retrieve 3D macromolecular structures and metadata" plus "structure discovery (text/sequence/3D similarity)" — which is a 4 ("lists several specific actions; minor gaps") on its own, but it uses second person ("use when you need"), which the guidelines penalize by reducing the specificity score by 1. It also omits formats (PDB/mmCIF) and batch retrieval as explicit actions.

3 / 5

Completeness

Both 'what' and 'when' are explicitly and concretely answered: what = "search, download, and programmatically retrieve 3D macromolecular structures and metadata"; when = "use when you need structure discovery (text/sequence/3D similarity) or automated structural data ingestion for structural biology and drug discovery workflows". This matches the anchor-5 example pattern of capability list followed by a concrete 'Use when...' clause.

5 / 5

Trigger Term Quality

Good natural keyword coverage: "RCSB Protein Data Bank (PDB)", "3D macromolecular structures", "structure discovery", "text/sequence/3D similarity", "structural biology", "drug discovery". A few natural terms users would say are missing (e.g., "protein structure", "PDB file", "mmCIF", "homolog"), so it fits anchor 4 rather than 5.

4 / 5

Distinctiveness Conflict Risk

The niche is unambiguous — RCSB Protein Data Bank access for structural biology and drug discovery — with domain-specific triggers ("3D similarity", "macromolecular structures") that no general-purpose skill would claim. Minimal conflict risk, matching anchor 5.

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/pdb-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.