extract

Extract structured data from websites and produce an executable Playwright script plus extracted data. Use when the user wants to scrape, extract, pull, collect, or harvest data from any website — product listings, tables, search results, feeds, profiles, or any repeating content.

Quality

88%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

77%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A dense, actionable extraction skill with strong code examples, a clear validated workflow, and useful error/selector-priority tables. Its main weaknesses are repeated routing guidance that bloats conciseness and a monolithic structure that does not progressively disclose the heavier mechanism/template material into separate reference files.

Suggestions

Consolidate the selector-routing logic (Decision Strategy diagram, the Non-negotiable rule, and Step 2/3 Paths A–C) into a single authoritative section to remove repetition and tighten conciseness.

Move the mechanism-aware code patterns (streaming/hydration, virtualized lists, infinite scroll, pagination) and the full script template into one-level-deep reference files (e.g. references/mechanisms.md, references/script-template.cjs) referenced from the overview, improving progressive disclosure.

Keep only a concise quick-start overview plus decision table inline so SKILL.md stays a lean entry point that signals the deeper material.

Dimension	Reasoning	Score
Conciseness	Mostly efficient and free of basic-concept padding, but the selector routing logic is restated three times (Decision Strategy diagram, the "Non-negotiable rule", and Execution Chain Step 2/3 Paths A–C) and could be consolidated; not the lean every-token-earns-its-place level of 3.	2 / 3
Actionability	Provides fully executable Playwright snippets (hydration, virtualized scroll, infinite scroll, pagination), concrete `actionbook` commands, a complete script template, and copy-paste-ready guardrails — matching the executable-and-complete anchor.	3 / 3
Workflow Clarity	A clearly sequenced 6-step chain (Understand → Obtain selectors → Probe → Generate → Execute & validate → Deliver) with an explicit validation table and a validate→fix→re-run feedback loop for the batch extraction operation.	3 / 3
Progressive Disclosure	Sections are well-organized, but the file is a ~340-line monolith with no bundle files; substantial content (mechanism-aware code patterns, full script template) that could live in one-level-deep reference files is kept inline, so it is not the clean overview-plus-references level of 3.	2 / 3
	Total	10 / 12 Passed

Description

100%

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, third-person description that pairs concrete capabilities with an explicit Use-when trigger clause and natural user vocabulary. No first/second-person voice, no vague fluff, and a distinct web-extraction niche.

Dimension	Reasoning	Score
Specificity	Names multiple concrete actions — "Extract structured data from websites", "produce an executable Playwright script", "plus extracted data" — matching the multi-action anchor rather than the single-domain anchor at 2.	3 / 3
Completeness	Explicitly answers both what (extract data + produce Playwright script) and when via a clear "Use when the user wants to..." clause, so it is not capped at 2 for a missing trigger clause.	3 / 3
Trigger Term Quality	Surfaces natural verbs users say — "scrape, extract, pull, collect, or harvest" — plus content nouns ("product listings, tables, search results, feeds, profiles"), giving broad coverage rather than the partial set at 2.	3 / 3
Distinctiveness Conflict Risk	A clear web-scraping niche with distinct triggers (harvest listings/tables/feeds/profiles) unlikely to collide with other skills, matching the distinct-niche anchor.	3 / 3
	Total	12 / 12 Passed

Validation

100%

Warnings & errors only

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

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: actionbook/actionbook
Path: skills/extract/SKILL.md
Commit: 4feb291

Reviewed: 3 days ago

Table of Contents

Discovery Implementation Validation

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.