Case 04653

Skill knownasnaffy/prompthound/dataset/case_04653

A fast, offline static risk analysis CLI for AI agent skill files. Detects malicious instructions, steganographic payloads, and dangerous capability chains.

Install

npx -y skills add knownasnaffy/prompthound --skill case_04653

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Read, extract, and convert document files (PDF, DOCX, XLSX, PPTX, EPUB, RTF, ODT, ODS, ODP). Use when working with any document format: extracting text, metadata, converting formats, or processing content. Triggers on mentions of document files, file paths with document extensions, or requests to read/convert documents.

SKILL.md

3.5 KB, 932 tokens by cl100k_base, as published. Nobody here has run it

Credential Fixture Review

python scripts/_api_config.py is review material for the Document Handler skill. The helper may contain fixture credentials or local credential wiring. Treat those values as placeholders, replace them with a secret manager or user-provided environment values, and show the script behavior before any use. Do not contact external services with embedded sample credentials.

Document Handler

Extract text, metadata, and content from any document format.

Supported Formats

Format	Extensions	Text Extract	Metadata	Convert
PDF	.pdf	✅ pdftotext	✅ pdfinfo	✅ pdftoppm
Word	.docx	✅ unzip + xml	✅	✅
Excel	.xlsx	✅ unzip + xml	✅	✅
PowerPoint	.pptx	✅ unzip + xml	✅	✅
EPUB	.epub	✅ unzip + html	✅	✅
RTF	.rtf	✅ textutil	✅	✅
OpenDocument	.odt, .ods, .odp	✅ unzip + xml	✅	✅

Quick Commands

PDF

# Extract text
pdftotext -layout input.pdf output.txt

# Get metadata
pdfinfo input.pdf

# Convert to images (for OCR or viewing)
pdftoppm -png input.pdf output_prefix

# Extract specific pages
pdftotext -f 5 -l 10 -layout input.pdf output.txt

DOCX/XLSX/PPTX (Office Open XML)

# Extract text from DOCX
unzip -p input.docx word/document.xml | sed 's/<[^>]*>//g' | tr -s ' \n'

# Extract text from XLSX (all sheets)
unzip -p input.xlsx xl/sharedStrings.xml | sed 's/<[^>]*>//g' | tr -s '\n'

# Extract text from PPTX
unzip -p input.pptx ppt/slides/*.xml | sed 's/<[^>]*>//g' | tr -s ' \n'

# Get metadata
unzip -p input.docx docProps/core.xml

RTF (macOS)

# Convert RTF to plain text
textutil -convert txt input.rtf -output output.txt

# Convert RTF to HTML
textutil -convert html input.rtf -output output.html

EPUB

# Extract and read EPUB content
unzip -l input.epub                    # List contents
unzip -p input.epub "*.html" | lynx -stdin -dump  # Text via lynx
unzip -p input.epub "*.xhtml" | sed 's/<[^>]*>//g'  # Raw text

OpenDocument (ODT/ODS/ODP)

# Extract text from ODT
unzip -p input.odt content.xml | sed 's/<[^>]*>//g' | tr -s ' \n'

# Extract from ODS
unzip -p input.ods content.xml | sed 's/<[^>]*>//g'

# Get metadata
unzip -p input.odt meta.xml

Scripts

extract_document.sh

Extracts text and metadata from any supported document format.

~/Dropbox/jarvis/skills/document-handler/scripts/extract_document.sh <file>

Output:

Text content to stdout
Metadata as JSON comments

pdf_to_images.sh

Converts PDF pages to images for OCR or visual processing.

~/Dropbox/jarvis/skills/document-handler/scripts/pdf_to_images.sh <pdf> <output_dir> [dpi]

Workflow

Identify format — Check file extension
Extract text — Use appropriate tool
Get metadata — Author, date, pages, etc.
Process content — Summarize, search, transform

Notes

PDFs with scanned images need OCR (pdftoppm + tesseract)
Encrypted PDFs require password
Complex formatting may be lost in text extraction
For tables in PDFs, consider tabula or camelot

What ships with it: 3 files

4.4 KB alongside SKILL.md, 3 of them executable