CtrlK
BlogDocsLog inGet started
Tessl Logo

prioritize-reference-files-6e4111

Always read and use provided reference files for data before attempting external searches or fabricating information

49

Quality

52%

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

Fix and improve this skill with Tessl

tessl review fix ./benchmarks/gdpval/skills/prioritize-reference-files-6e4111/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

62%

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The content is well-structured and self-contained with concrete DOCX fallback code, but it is padded with redundant anti-pattern/checklist restatements, leans on pseudocode for the primary read actions, and lacks explicit validation checkpoints in the main data-extraction workflow. Progressive disclosure is fine given it is a simple single-file skill.

Suggestions

Tighten conciseness by collapsing the Anti-Patterns and Checklist sections into the Steps, since they restate the same guidance; cut the 'Core Principle' paragraph that repeats the frontmatter idea.

Replace the pseudocode read_file(filetype="xlsx") examples with real, executable snippets (e.g., openpyxl/pandas for xlsx, csv module for csv) so the primary path is copy-paste ready like the DOCX fallback.

Add an explicit validate-data checkpoint in the main flow (e.g., verify row counts/columns and flag gaps before populating output) to close the workflow-clarity validation gap.

DimensionReasoningScore

Conciseness

The body is mostly actionable but padded with redundant restatements (the Anti-Patterns and Checklist largely restate the Steps) and verbose explanations of an obvious principle, so it could be tightened considerably despite being accurate.

2 / 3

Actionability

It provides concrete, copy-pasteable fallback code for the DOCX edge case and a decision-flow snippet, but the core read_file examples use non-executable pseudocode signatures (filetype="xlsx") and the main steps (Steps 3-4) give abstract direction rather than runnable commands.

2 / 3

Workflow Clarity

Steps are clearly sequenced (1-5) with a checklist, but there is no validation checkpoint in the main flow; only the DOCX fallback mentions 'Verify the extracted content is complete and readable,' and a data-extraction workflow with no verify step is a validation gap that caps clarity at 2.

2 / 3

Progressive Disclosure

No references/scripts/assets bundle exists, and the SKILL.md body is a single self-contained, well-organized file with clear section headers; per the rubric's simple-skill note, a skill under 50 lines with no need for external references can score 3 with well-organized sections.

3 / 3

Total

9

/

12

Passed

Description

42%

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 conveys the core principle (prioritize provided reference files over fabricated/external data) but is vague, lacks a 'Use when' activation trigger, and is not distinctive enough to avoid conflicting with other file-handling skills. It reads as a behavioral rule rather than an activatable skill description.

Suggestions

Add an explicit 'Use when...' trigger clause naming natural user phrases like 'provided files', 'attached spreadsheet', or 'uploaded document'.

List concrete actions the skill performs (e.g., extract data from .xlsx/.csv/.pdf/.docx, validate content, cite sources) to raise specificity above a single generic behavior.

Sharpen the niche so it is distinguishable from general file-handling skills, e.g. 'Use when a task ships structured data files (xlsx/csv/pdf/docx) and the agent must source output values from them instead of generating or web-searching data.'

DimensionReasoningScore

Specificity

The description names a concrete behavior ("read and use provided reference files for data") but stops at a single generalized action rather than listing multiple specific actions like "extract," "validate," or "cite" data.

2 / 3

Completeness

It states clearly what the skill does (read/prioritize reference files) but offers no "Use when..." trigger clause or equivalent guidance on when Claude should activate it, which the rubric caps completeness at 2.

2 / 3

Trigger Term Quality

It includes some relevant terms ("reference files," "data," "external searches"), but lacks the natural variations a user would say (e.g., "attached file," "provided spreadsheet," "uploaded document"), and "fabricating information" is internal jargon rather than a user phrase.

2 / 3

Distinctiveness Conflict Risk

"Read and use provided reference files" is extremely generic and would overlap with many file-handling skills, with no distinctive trigger to disambiguate when this skill applies versus others.

1 / 3

Total

7

/

12

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
HKUDS/OpenSpace
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.