file-reading
Use this skill when a file has been uploaded but its content is NOT in your context — only its path at /mnt/user-data/uploads/ is listed in an uploaded_files block. This skill is a router: it tells you which tool to use for each file type (pdf, docx, xlsx, csv, json, images, archives, ebooks) so you read the right amount the right way instead of blindly running cat on a binary. Triggers: any mention of /mnt/user-data/uploads/, an uploaded_files section, a file_path tag, or a user asking about an uploaded file you have not yet read. Do NOT use this skill if the file content is already visible in your context inside a documents block — you already have it.
94
92%
Does it follow best practices?
Impact
—
No eval scenarios have been run
Advisory
Suggest reviewing before use
Quality
Discovery
100%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This is a strong, well-crafted skill description that clearly defines its role as a file-type router for uploaded but unread files. It excels in all dimensions: specific actions and file types are enumerated, natural trigger terms are comprehensive, both 'what' and 'when' are explicitly addressed, and the negative condition creates a clear boundary. The only minor issue is the use of second person ('you') throughout, but since this is an internal routing skill addressed to Claude itself, this is contextually appropriate and doesn't reduce clarity.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple concrete actions and file types: routing to the right tool for pdf, docx, xlsx, csv, json, images, archives, ebooks. Clearly describes what it does ('tells you which tool to use for each file type') and contrasts with the wrong approach ('instead of blindly running cat on a binary'). | 3 / 3 |
Completeness | Clearly answers both 'what' (a router that tells which tool to use for each file type) and 'when' (explicit triggers including path mentions, uploaded_files blocks, file_path tags, and user asking about unread uploads). Also includes a clear 'do NOT use when' clause for when content is already in context. | 3 / 3 |
Trigger Term Quality | Excellent coverage of natural trigger terms: '/mnt/user-data/uploads/', 'uploaded_files section', 'file_path tag', 'uploaded file you have not yet read', plus specific file type keywords (pdf, docx, xlsx, csv, json, images, archives, ebooks). These are terms that would naturally appear in context when this skill is needed. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive with a clear niche: it specifically handles the case of uploaded files whose content is not yet in context, acting as a routing layer. The explicit negative condition ('Do NOT use if file content is already visible') further reduces conflict risk with other file-handling skills. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
85%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a strong, well-structured routing skill that provides concrete, executable guidance for every common file type. The dispatch table is an excellent organizational choice, and the progressive disclosure to dedicated skills is clean and well-signaled. Minor verbosity in the introductory section and some edge-case explanations could be trimmed, but overall the content is highly actionable and well-organized.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is mostly efficient and well-targeted, but includes some unnecessary explanations Claude would already know (e.g., explaining what binary garbage is, explaining RFC-4180 quoted newlines, explaining that 'data' from `file` means unrecognized). The 'Why this skill exists' section could be trimmed. However, most content earns its place as dispatch-specific guidance. | 2 / 3 |
Actionability | Every file type has concrete, executable commands and code snippets that are copy-paste ready. The dispatch table provides specific first moves, the code examples use real libraries with correct syntax, and edge cases (legacy formats, read_only mode, engine parameters) are addressed with working code. | 3 / 3 |
Workflow Clarity | The general protocol establishes a clear sequence (check extension → stat before read → read just enough → hand off to dedicated skill). Each file type section follows a consistent pattern of safe first move → deeper exploration → when to escalate. Validation checkpoints are present (e.g., stat before read, wc -l before full load, list archives before extracting). | 3 / 3 |
Progressive Disclosure | The skill is structured as a router/overview with a clear dispatch table pointing to dedicated skills (pdf-reading, docx, xlsx, pptx) via explicit file paths. Inline content covers only the quick-peek first move, with clear handoff signals like 'go read /mnt/skills/public/pdf-reading/SKILL.md' for deeper operations. Navigation is one level deep and well-signaled. | 3 / 3 |
Total | 11 / 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.
Validation — 11 / 11 Passed
Validation for skill structure
No warnings or errors.
- Repository
- douglasvought/wiggle-skills
- Commit
b27906e
- 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.