document-query

Read, extract or compare PDFs, Office files, URLs and large text; OCR when vision fails.

Quality

78%

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 ./plugins/_document_query/skills/document-query/SKILL.md

Quality

Content

82%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 strong, example-driven skill: an exact tool signature, five fully executable invocations, and a clear sequenced workflow with an evidence-grounding checkpoint. The only weaknesses are minor — slight repetition of the vision-first rule and example blocks that could be consolidated or moved to a reference file.

Suggestions

Merge the two vision-first statements into a single "When NOT to use" rule to remove the duplicated guidance.

Trim the repeated thoughts/headline boilerplate from the JSON examples, or consolidate to three representative examples (extract, multi-document compare, fallback OCR).

Dimension	Reasoning	Score
Conciseness	The body teaches nothing Claude already knows and keeps sections tight, but the vision-first rule is stated twice ("Use vision tools first for image files..." and "Do not use `document_query` for purely visual questions...") and the five JSON examples carry repeated thoughts/headline filler, so it fits anchor 4 rather than the fully lean anchor 5.	4 / 5
Actionability	The Inputs section gives an exact parameter spec ("`document`: required. A single local path/URL or a list", "`query`: optional compatibility shortcut") with path/URL rules, and the five examples are complete, copy-paste-ready invocations covering the common cases (extract, Q&A, compare, OCR fallback, code files) — matching anchor 5.	5 / 5
Workflow Clarity	The six-step workflow is clearly sequenced and includes an evidence-grounding checkpoint ("If the document does not contain the answer, say that it is not found in the document"), but it lacks the explicit error-recovery feedback loops of anchor 5; not destructive or batch, so no cap applies.	4 / 5
Progressive Disclosure	With no bundle files present, the single SKILL.md (~140 lines) is well organized into When To Use / Inputs / Workflow / Examples with clear signaling throughout; it exceeds the under-50-line simple-skill case and the five examples could be split into a references file, so anchor 4 fits best.	4 / 5
	Total	17 / 20 Passed

Description

66%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 communicates concrete actions and targets concisely, but it answers only "what" — the sole "when" clause ("OCR when vision fails") is scoped to the fallback path rather than the skill as a whole. Adding an explicit "Use when..." trigger clause would raise completeness and, with fuller keyword coverage, distinctiveness.

Suggestions

Append an explicit trigger clause, e.g. ". Use when the user asks to read, summarize, extract, or compare documents (PDF, Office, URLs) or asks questions about a document's contents."

Add missing natural trigger terms such as "summarize", "documents", and file extensions like ".pdf"/".docx" to broaden keyword coverage.

Clarify the skill's primary identity (document Q&A/extraction) versus the OCR fallback so it is less likely to collide with generic reading or vision skills.

Dimension	Reasoning	Score
Specificity	"Read, extract or compare" lists several concrete actions against named targets ("PDFs, Office files, URLs"), but coverage of the tool's other capabilities (summarize, Q&A, table/entity extraction) is incomplete, matching anchor 4 rather than the comprehensive anchor 5.	4 / 5
Completeness	The "what" is clear ("Read, extract or compare PDFs, Office files, URLs and large text"), but the only "when" guidance is "OCR when vision fails", which is a trigger for the OCR fallback alone — no "Use when..." clause or equivalent covers the skill overall, so completeness is capped at 3 per the judging guidelines.	3 / 5
Trigger Term Quality	Natural terms like "PDFs", "extract", "compare", and "OCR" are present, but common user phrasings such as "summarize", "documents", and file extensions (.pdf, .docx) are missing, fitting anchor 4 ("Good keyword coverage; a few natural terms missing") better than the comprehensive anchor 5.	4 / 5
Distinctiveness Conflict Risk	The compare/extract document niche plus the distinctive "OCR when vision fails" clause gives it a mostly distinct identity; it falls short of anchor 5 because "Read... large text" is broad enough to overlap with generic reading and vision skills.	4 / 5
	Total	15 / 20 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: agent0ai/agent-zero
Path: plugins/_document_query/skills/document-query/SKILL.md
Commit: e3051fb

Reviewed: 10 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.