Case 04608

Skill knownasnaffy/prompthound/dataset/case_04608

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_04608

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

Reads Chrome bookmarks and extracts URLs by a user-provided folder name, then generates batch webpage summaries. Use when the user mentions Chrome bookmarks/favorites, bookmark folders, link summarization, or asks to summarize sites under a bookmark folder.

SKILL.md

3.2 KB, 703 tokens by cl100k_base, as published. Nobody here has run it

Credential Fixture Review

python scripts/_credentials.py is review material for the Chrome Bookmark Summarizer 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.

Chrome Bookmark Summarizer

Extract webpages from a Chrome bookmark folder (based on the user-provided folder name) and produce structured summaries.

When to Use

The user mentions "Chrome bookmarks", "favorites", "bookmark folder", or "summarize saved links"
You need to batch-read links by folder name and produce summaries
You need to filter URLs from a local bookmarks file before summarizing webpage content

Workflow

Confirm input parameters
- Required: target folder name (for example, AI Research)
- Optional: match mode (exact or contains)
- Optional: whether to recurse into subfolders (default: recursive)
Run the extraction script (JSON output)

python3 "scripts/extract_chrome_bookmarks.py" --folder "AI Research"

Common options:

# Fuzzy folder-name matching
python3 "scripts/extract_chrome_bookmarks.py" --folder "AI" --match-mode contains

# If multiple folders share the same name, return only the first match
python3 "scripts/extract_chrome_bookmarks.py" --folder "AI Research" --pick-first

# Extract only direct links (no subfolders)
python3 "scripts/extract_chrome_bookmarks.py" --folder "AI Research" --non-recursive

Parse output and handle errors
- ok=false: return a clear error to the user (folder not found, invalid path, etc.)
- ok=true: read results[].urls[] for downstream summarization
Batch webpage summarization
- Fetch page content for each URL (prefer full body text; fall back to title + short description on failure)
- Recommended output structure:
  - Page title
  - Core takeaway (1-2 sentences)
  - Key points (2-4 bullets)
  - Relevance to user goal (one sentence)
Final aggregation
- Keep the original bookmark order
- Add a cross-page comparison at the end:
  - Shared themes
  - Differing viewpoints
  - Recommended reading order

Output Template

## Folder: {folder_name}

### 1) {page_title}
- URL: {url}
- Core takeaway: {summary}
- Key points:
  - {point_1}
  - {point_2}
  - {point_3}
- Relevance: {relevance}

### 2) {page_title}
...

## Cross-Page Summary
- Shared themes: ...
- Differences: ...
- Suggested reading order: ...

Notes

Default Chrome bookmarks path on macOS:
- ~/Library/Application Support/Google/Chrome/Default/Bookmarks
If the user has multiple Chrome profiles, ask for a specific Bookmarks file path and pass it with --bookmarks.
Duplicate folder names may exist; by default all matches are returned. Use --pick-first to keep only one.