pdf-extraction-fallbacks

Multi-fallback PDF/text extraction with early failure detection and sequential tool fallbacks

Quality

63%

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

Fix and improve this skill with Tessl

tessl review fix ./benchmarks/gdpval/skills/pdf-extraction-fallbacks/SKILL.md

Quality

Content

77%

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 strong, actionable extraction workflow with explicit validation checkpoints and a complete example script, but it is somewhat verbose due to duplicated logic and inline content that could be split into referenced files.

Suggestions

Move the full 'Example Complete Workflow Script' and 'Common Failure Modes' table into referenced files (e.g. references/extract_pdf_robust.sh, references/failure_modes.md) and link them from the main body to improve progressive disclosure and reduce token cost.

Remove inline commentary that restates tool capabilities ('fastest, most reliable', 'better for tables and structured content') since Claude already knows these libraries.

Consolidate the duplicated validation logic between the step-by-step examples and the final script so the guidance is stated once.

Dimension	Reasoning	Score
Conciseness	Mostly efficient with executable code and tight tables, but it includes commentary Claude doesn't need (e.g. 'fastest, most reliable', 'better for tables') and duplicates validation logic across the inline steps and the full example script.	2 / 3
Actionability	Each step provides executable bash/Python with concrete thresholds and a copy-paste-ready full script, satisfying the 'fully executable; copy-paste ready' anchor despite a couple of comment-only fallback notes.	3 / 3
Workflow Clarity	A clear numbered sequence (Steps 1-6) with an explicit decision tree, validation checkpoints (size, magic bytes, word/char thresholds), and feedback loops ('try next' on failure).	3 / 3
Progressive Disclosure	Well-organized into clear sections but everything is inline in a single ~257-line file, with content that could be split (the full example script, failure-mode table) not separated or referenced.	2 / 3
	Total	10 / 12 Passed

Description

50%

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 clearly communicates a focused PDF-extraction niche and a concrete strategy, but it omits the 'when to use it' trigger guidance and relies on somewhat technical phrasing rather than natural user terms.

Suggestions

Add an explicit 'Use when ...' clause naming natural triggers, e.g. 'Use when extracting text from PDFs that may be protected, corrupted, or JavaScript-restricted, or when a single extraction tool returns empty/garbled output.'

Replace internal jargon like 'sequential tool fallbacks' with user-facing terms such as 'try pdftotext, PyMuPDF, and pdfplumber in order' so users naturally trigger the skill.

List discrete concrete actions (extract text, detect corrupt/error-page downloads, handle protected PDFs) to lift specificity beyond a strategy description.

Dimension	Reasoning	Score
Specificity	Names the domain and several specific concepts ('multi-fallback PDF/text extraction', 'early failure detection', 'sequential tool fallbacks'), but describes a strategy rather than enumerating discrete concrete actions.	2 / 3
Completeness	States what the skill does but provides no 'Use when...' or equivalent explicit trigger guidance, so the 'when' is only implied, which caps completeness at 2.	2 / 3
Trigger Term Quality	Contains a real natural term ('PDF/text extraction') but misses common variations users would say ('PDFs', 'extract text', 'parse PDF') and leans on technical phrasing ('sequential tool fallbacks').	2 / 3
Distinctiveness Conflict Risk	The multi-fallback-with-failure-detection niche is fairly specific, but without explicit trigger scoping it could still overlap with other generic PDF-extraction skills.	2 / 3
	Total	8 / 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: HKUDS/OpenSpace
Path: benchmarks/gdpval/skills/pdf-extraction-fallbacks/SKILL.md
Commit: c5a9c4b

Reviewed: 2 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.