yao-ocr

OCR text recognition expert. ALWAYS invoke this skill when you need to extract text from images or PDFs — including invoices, receipts, ID cards, bank cards, business licenses, tables, handwritten documents, or any visual text content.

Quality

90%

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

90%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 a tight, executable reference: parameter and type tables plus concrete command examples make it highly actionable, and the simple single-purpose scope fits cleanly in one well-structured file. The only minor inefficiency is partial duplication between the Guidelines and the tables above.

Suggestions

Trim the 'Guidelines' bullets that restate the parameter/provider tables (e.g. PDF-support and prompt-only-with-VLM notes already appear in the tables) to remove redundancy.

Dimension	Reasoning	Score
Conciseness	Largely lean and tabular with executable commands and no padding about what OCR/PDFs are; the 'Guidelines' section partly repeats information already covered in the parameter and provider tables, which could be trimmed.	4 / 5
Actionability	Copy-paste-ready `tai tool ocr_recognize` commands cover the common cases (plain text, URL, table, invoice, provider, VLM prompt, page range), backed by a full parameter table and recognition-type table.	5 / 5
Workflow Clarity	This is a single-purpose skill where the action is unambiguous (run one command with documented flags); the simple-skill exception applies, and the Guidelines section sequences provider selection and output-format choices clearly.	5 / 5
Progressive Disclosure	Well-organized single-file skill under 50 lines with no external references needed; content is appropriately sectioned (tools, parameters, types, PDF support, multi-page response, guidelines), so the no-bundle exception applies.	5 / 5
	Total	19 / 20 Passed

Description

90%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.

The description is strong: it explicitly states what the skill does and when to invoke it, with rich natural trigger terms covering many document types. Third-person voice is maintained. Minor risk comes from the very broad 'any visual text content' catch-all.

Suggestions

Tighten the catch-all 'any visual text content' so it does not invite triggering for non-OCR image-analysis tasks unrelated to text recognition.

Dimension	Reasoning	Score
Specificity	Names the OCR domain and several concrete actions ('extract text from images or PDFs', structured invoice/receipt/id/bank/license extraction, tables, handwriting); coverage is broad but the phrasing collapses many actions into a single 'extract' verb with a long enumeration, leaving minor gaps.	4 / 5
Completeness	Explicitly answers both 'what' (OCR text recognition, extracting text from images/PDFs and listed document types) and 'when' (an explicit 'ALWAYS invoke this skill when you need to extract text from images or PDFs — including...').	5 / 5
Trigger Term Quality	Comprehensive natural terms users would say — 'images or PDFs', 'invoices, receipts, ID cards, bank cards, business licenses, tables, handwritten documents' — with strong synonym coverage and concrete document-type triggers.	5 / 5
Distinctiveness Conflict Risk	Clear OCR niche with concrete document-type triggers makes it largely distinct, but the broad 'any visual text content' phrasing creates minor overlap risk with general image/vision skills.	4 / 5
	Total	18 / 20 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: YaoApp/yao
Path: tools/skills/yao-ocr/SKILL.md
Commit: d90d41d

Reviewed: 21 hours 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.