CtrlK
BlogDocsLog inGet started
Tessl Logo

sci-data-extractor

Extract structured data from scientific literature PDFs using AI-powered OCR and LLM analysis. Supports enzyme kinetics, experimental results, and literature review templates. Use when researchers need to parse tables, charts, or text from paper PDFs into Markdown or CSV. Triggers: extract data from PDF, scientific data extraction, parse paper tables, enzyme kinetics extraction, batch PDF extraction, literature data, research data from paper, OCR scientific paper, convert PDF to structured data.

68

Quality

85%

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

SKILL.md
Quality
Evals
Security

Quality

Content

82%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, highly actionable skill body: every workflow step is an executable command, and failure modes get concrete WHY/BAD/GOOD treatment. The main improvements are integrating validation checkpoints into the workflow steps rather than isolating them in Anti-Patterns, trimming the 'gotcha' narration, and fixing the dead requirements.txt reference.

Suggestions

Inline the validation checkpoints into the workflow steps (e.g. after step 7, add 'review the final summary printed by batch_extract.py and re-run failed files with --print') so the sequence is self-contained.

Remove the dead [requirements.txt](requirements.txt) link from the References section, or add the file to the bundle.

Tighten the Mindset section and Anti-Pattern WHYs by dropping the repeated 'a production gotcha is...' framing — state the failure mode once in plain terms.

DimensionReasoningScore

Conciseness

Commands are lean and copy-paste ready, and setup detail is correctly delegated to the reference file, but there is trimmable phrasing — the Mindset section's "a gotcha is assuming..." framing and the repeated "a production gotcha is..." constructions across Anti-Patterns add words without adding information. Not 5 due to these minor instances of over-explanation; not 3 because nothing explains concepts Claude already knows and no section is padded.

4 / 5

Actionability

Every workflow step is a fully executable command with real flags and outputs — "./scripts/extractor.py paper.pdf --template enzyme -o results.md", "--format csv", "--ocr mathpix", "-p" custom prompt example, "batch_extract.py ./literature/ ./output/ --template enzyme --format csv", and "--print". Covers the common cases (presets, custom prompt, CSV, OCR, batch) copy-paste ready. Not 4 because there are no gaps in executable detail.

5 / 5

Workflow Clarity

The eight-step sequence is clearly ordered and validation is genuinely documented (spot-check 3 rows against the source PDF, read the batch failed-files summary, check extracted text length before switching OCR), so the batch-operation cap does not apply. Not 5 because those checkpoints live in the Anti-Patterns section rather than being integrated into the workflow steps themselves — e.g. step 7 (batch) doesn't inline 'then review the final summary'.

4 / 5

Progressive Disclosure

Good structure: the body stays a readable overview, setup/troubleshooting is properly pushed to a real one-level-deep reference (references/setup-and-troubleshooting.md exists and is linked inline and in a References section), and scripts exist as referenced. Not 5 because the References section links [requirements.txt](requirements.txt) which is absent from the bundle — a dead reference — a minor organization gap fitting the level-4 anchor.

4 / 5

Total

17

/

20

Passed

Description

88%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 that clearly states capabilities, output formats, and explicit use-when triggers with a natural keyword list. Its only weaknesses are a few missing trigger synonyms (file extensions, figure-related terms) and mildly generic PDF-extraction phrases that could overlap with general PDF skills.

DimensionReasoningScore

Specificity

Lists multiple concrete actions — "parse tables, charts, or text from paper PDFs into Markdown or CSV", "AI-powered OCR and LLM analysis", plus named template domains (enzyme kinetics, experimental results, literature review) — giving comprehensive coverage of the skill's capabilities. Not 4 because coverage goes beyond 'several actions with minor gaps': extraction, conversion, OCR, and batch processing are all named.

5 / 5

Completeness

Explicitly answers both questions: the 'what' ("Extract structured data from scientific literature PDFs... into Markdown or CSV") and a concrete 'when' ("Use when researchers need to parse tables, charts, or text from paper PDFs") reinforced by a nine-phrase trigger list. Not 4 because the 'when' clause is specific and concrete rather than merely adequate.

5 / 5

Trigger Term Quality

The trigger list is strong and natural ("extract data from PDF", "parse paper tables", "batch PDF extraction", "OCR scientific paper", "convert PDF to structured data"), but a few natural terms are missing — no ".pdf" file extension and no figure/graph synonyms for "charts". Not 5 because the top anchor expects synonyms plus file extensions; clearly above 3 since common variations are well covered.

4 / 5

Distinctiveness Conflict Risk

The scientific-literature niche with domain triggers ("enzyme kinetics extraction", "OCR scientific paper") is mostly distinct, but generic phrases like "extract data from PDF" and "convert PDF to structured data" create minor overlap risk with general PDF-extraction skills. Not 5 because of that residual overlap; not 3 because the dominant triggers are domain-specific.

4 / 5

Total

18

/

20

Passed

Validation

81%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation13 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

allowed_tools_field

'allowed-tools' contains unusual tool name(s)

Warning

metadata_version

'metadata.version' is missing

Warning

relative_links

Relative link issues: 1 missing

Warning

Total

13

/

16

Passed

Repository
pantheon-org/tekhne
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.