CtrlK
BlogDocsLog inGet started
Tessl Logo

pdf-skill

PDF 读/写/转/OCR 一体化。触发场景:总结 PDF、读文本或表格、填 AcroForm 表单、渲染页面为图、合并/拆分/水印/旋转/加密、创建新 PDF(发票/报告)、PDF ↔ docx/md/txt 互转。典型说法:“总结这份 PDF”、“填一下这个 PDF 表单”、“把几个 PDF 合一起”、“PDF 加水印”、“PDF 转 Word”、“扫描件 OCR”、“看看 PDF 第 3 页长什么样”。用户提到 .pdf 文件、要产出 PDF、或要对 PDF 做任何改动都加载本技能。

76

Quality

95%

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

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 dense, executable reference: lean tables and copy-paste-ready pdf-lib/pdfkit demos with a clear read→write→OCR→convert flow and a troubleshooting feedback table. Its main gap is the absence of explicit output-validation checkpoints on destructive/batch write operations, which holds workflow clarity just below the top anchor.

Suggestions

Add an explicit validation checkpoint after destructive/batch write demos (e.g., re-open the output with PdfInspect summary or verify page count/opens correctly) before declaring success, to lift workflow clarity to 5.

Consider extracting the five demo code blocks into a references/ file (e.g. references/demos.md) with one-level-deep links from SKILL.md to improve progressive disclosure for the larger body.

Make the read-branch encryption path explicit as a sequenced sub-step (detect → supply password or JsSandbox decrypt → retry) rather than an inline table cell, to tighten the feedback loop on encrypted files.

DimensionReasoningScore

Conciseness

Lean and table-driven with terse command snippets and executable demo code; the few prose notes (CJK font registration, pdfkit save() gotcha, 兜底 table) are environment-specific knowledge Claude would not infer, so every token earns its place. Not below because there is no padding or explanation of concepts Claude already knows.

5 / 5

Actionability

Fully executable, copy-paste-ready code blocks for invoice creation, watermarking, merge, form-fill, decrypt/rotate/extract plus concrete tool invocations (PdfInspect {action,filePath}, DocConvert(from,to,sourcePath), CloudImageUnderstand(...)) covering the common cases. Not below because examples are complete with imports and file writes, not pseudocode.

5 / 5

Workflow Clarity

Clear sequencing with a read decision tree (textType → 下一步 branching table), a 3-step OCR pipeline, and a symptom→cause→fix 兜底 table plus an edit-and-run retry path acting as a feedback loop. Not a 5 because the destructive/batch write demos (merge, watermark, decrypt, rotate) finalize with only console.log and lack an explicit output-validation checkpoint; not a 3 because error-recovery feedback loops are present.

4 / 5

Progressive Disclosure

Single self-contained file with clear section headers (工具清单, numbered 1–5) and no nested or buried references; content is appropriately inline for a tooling skill. Not a 5 because at ~200 lines it is monolithic with no one-level-deep reference split for the five demo blocks; not below because structure is clean and navigation is easy.

4 / 5

Total

18

/

20

Passed

Description

100%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 exemplary: it concisely states what the skill does, enumerates comprehensive concrete actions, and gives explicit natural-language trigger phrases plus the .pdf extension for when to load it. It is clearly distinguishable from sibling skills and uses appropriate directive voice.

DimensionReasoningScore

Specificity

Lists multiple concrete actions covering the full PDF lifecycle — "总结 PDF、读文本或表格、填 AcroForm 表单、渲染页面为图、合并/拆分/水印/旋转/加密、创建新 PDF…互转…OCR" — matching the comprehensive-coverage anchor; not below because coverage is broad and specific, and there is no higher anchor.

5 / 5

Completeness

Explicitly answers both what ("PDF 读/写/转/OCR 一体化" with concrete actions) and when ("触发场景…典型说法…用户提到 .pdf 文件…都加载本技能") with concrete trigger phrases, matching the top anchor; not below because neither side is implicit.

5 / 5

Trigger Term Quality

Comprehensive natural terms including synonyms and the .pdf extension: quoted user phrases ("总结这份 PDF"、"PDF 加水印"、"PDF 转 Word"、"扫描件 OCR") plus "用户提到 .pdf 文件", matching the top anchor; not below because both everyday phrasings and the file extension are present.

5 / 5

Distinctiveness Conflict Risk

Clear PDF-only niche with distinct, domain-specific triggers and file extension, giving minimal conflict risk; not below because the scope is sharply bounded to PDF operations rather than overlapping broad document skills.

5 / 5

Total

20

/

20

Passed

Validation

100%

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

Validation16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository
itsablabla/openloaf-web
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.