CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

pdf-processing-pro

综合办公文员与软件开发工程师当需要批量处理PDF表单、提取表格或进行OCR识别时,使用内置脚本一键完成自动化提取与数据校验,彻底告别繁琐的手动录入,让复杂文档工作流高效、稳健落地。

Quality

69%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./skills/pdf-processing-pro/pdf-processing-pro/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

52%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 well-structured with concrete, copy-paste-ready commands and clearly sequenced workflows that include validation steps. Its critical failure is that the bundle does not back the documentation: all three referenced reference files and 7 of 8 referenced scripts are missing, which breaks both navigation and executability, and marketing-flavored sections (Features, Best practices, Performance tips) pad the token budget.

Suggestions

Ship the referenced files: add FORMS.md, TABLES.md, and OCR.md (or remove those references), and include the seven documented scripts (fill_form.py, validate_form.py, extract_tables.py, extract_text.py, merge_pdfs.py, split_pdf.py, validate_pdf.py) or delete their sections.

Trim the Features, Best practices, and Performance tips sections to the non-obvious specifics (e.g., actual exit-code contract, the --parallel flag support matrix) instead of generic engineering advice.

Add a fix-and-retry loop to the workflows (e.g., 'if validate_form.py exits 4, correct the JSON fields it lists and re-run') and a requirements.txt since the troubleshooting section tells users to pip install -r requirements.txt.

DimensionReasoningScore

Conciseness

The body is mostly command/code listings with no concept explanations, but it carries unnecessary padding: the "Features" section markets generic script qualities ("Type hints: Full type annotations for IDE support", "CLI interface: --help flag"), and "Best practices"/"Performance tips" restate knowledge Claude already has ("Use try-except in custom scripts", "Set timeouts"). This fits the level-3 anchor (mostly efficient with some unnecessary explanation), not level 4 whose over-explanation would be only minor.

3 / 5

Actionability

The written guidance is concrete — copy-paste commands with flags ("python scripts/fill_form.py input.pdf data.json output.pdf") and executable pdfplumber snippets — but the actual bundle contains only scripts/analyze_form.py, so 7 of the 8 documented scripts (fill_form.py, validate_form.py, extract_tables.py, extract_text.py, merge_pdfs.py, split_pdf.py, validate_pdf.py) and requirements.txt do not exist and those commands fail as written. Concrete-but-incomplete matches the level-3 anchor; the gaps are too large for level 4's 'minor gaps'.

3 / 5

Workflow Clarity

The three workflows are clearly sequenced with numbered steps, and validation checkpoints are present (Workflow 1 runs validate_form.py then validate_pdf.py; the batch workflow checks returncode and reports stderr), so the batch-operation cap at 3 does not apply and it fits the level-4 anchor. It falls short of level 5 because there is no explicit fix-and-retry feedback loop — the error-handling section shows exit-code branching but never instructs how to recover and re-run.

4 / 5

Progressive Disclosure

Scored against the actual bundle: the body's section structure and one-level-deep signaling are good on paper, but every detailed-reference target is dead — FORMS.md, TABLES.md, and OCR.md (each referenced twice) do not exist, nor do 7 of the 8 linked scripts — so navigation points at nothing and content promised as 'separate' was never delivered. This is effectively broken organization rather than merely imperfect structure, placing it at the level-2 anchor rather than level 3's 'references present but not clearly signaled'.

2 / 5

Total

12

/

20

Passed

Description

75%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 answers both what and when with several concrete, natural trigger phrases and stays clearly within the PDF-processing niche. Its main weakness is marketing fluff ("彻底告别繁琐的手动录入", "高效、稳健落地") that pads the what-statement without adding capability information, and missing common variations like ".pdf" or "PDF文件".

Suggestions

Remove the marketing filler ("彻底告别繁琐的手动录入,让复杂文档工作流高效、稳健落地") and keep only capability statements.

Add missing natural trigger variations such as ".pdf", "PDF文件", and "合并/拆分PDF" to broaden keyword coverage.

Lead with the actions in third person (e.g., "批量处理PDF表单、提取表格、OCR识别与数据校验") before the when-clause, so the what is not buried after the audience framing.

DimensionReasoningScore

Specificity

The description names several concrete actions — "批量处理PDF表单" (batch process PDF forms), "提取表格" (extract tables), "OCR识别" (OCR recognition), "自动化提取与数据校验" (automated extraction and data validation) — which goes beyond the 1-2 actions of the level-3 anchor, but coverage has gaps (no text extraction, merge/split, file conversion) so it falls short of the comprehensive level-5 anchor.

4 / 5

Completeness

Both parts are present: an explicit when-clause ("当需要批量处理PDF表单、提取表格或进行OCR识别时") and a what-statement ("使用内置脚本一键完成自动化提取与数据校验"), so it clears the level-3 cap for missing triggers. However, the what is diluted by marketing padding ("彻底告别繁琐的手动录入,让复杂文档工作流高效、稳健落地") and the triggers are need-based rather than the concrete user-mention phrases of the level-5 anchor.

4 / 5

Trigger Term Quality

Good natural keyword coverage: "批量处理PDF表单", "提取表格", "OCR识别", "数据校验" are phrases users would plausibly say, matching the level-4 anchor, but common variations and file extensions (".pdf", "PDF文件", "合并/拆分") are missing, so it does not reach the level-5 anchor's comprehensive synonym/extension coverage.

4 / 5

Distinctiveness Conflict Risk

The PDF-specific triggers (PDF forms, PDF table extraction, OCR) carve out a mostly distinct niche with minor overlap risk against closely related document/table skills, fitting the level-4 anchor; it is not a level-5 because "提取表格" and generic validation wording could also fire for non-PDF table or data-validation tasks.

4 / 5

Total

16

/

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.

Validation — 13 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

relative_links

Relative link issues: 6 missing

Warning

referenced_paths_exist

Referenced path issues: 17 missing

Warning

Total

13

/

16

Passed

Repository

anbeime/skill

Path

skills/pdf-processing-pro/pdf-processing-pro/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.