tdg-personal/data-scraper-agent
Build a fully automated AI-powered data collection agent for any public source — job boards, prices, news, GitHub, sports, anything. Scrapes on a schedule, enriches data with a free LLM (Gemini Flash), stores results in Notion/Sheets/Supabase, and learns from user feedback. Runs 100% free on GitHub Actions. Use when the user wants to monitor, collect, or track any public data automatically.
64
81%
Does it follow best practices?
Run evals on this skill
Adds up to 20 points to the overall score
View guide
Low
Low-risk findings worth noting
Low
Low-risk findings.
1 low severity finding. Worth noting, but not necessarily harmful.
W011: Third-party content exposure detected (indirect prompt injection risk).
What this means
The skill exposes the agent to untrusted, user-generated content from public third-party sources, creating a risk of indirect prompt injection. This includes browsing arbitrary URLs, reading social media posts or forum comments, and analyzing content from unknown websites.
Why it was flagged
The skill's SKILL.md and code (e.g., scraper/sources/my_source.py and the HTML/RSS/Playwright patterns, plus ai/pipeline.py which builds prompts from scraped items and sends them to Gemini) explicitly fetch and ingest public/untrusted web content (examples include Hacker News, subreddits, LinkedIn, etc.) and uses the LLM's analyses to score/filter/store items, so third-party content can materially influence agent decisions.
Where we found it
arbitrary public web content (HTML, RSS, REST API, JS-rendered pages)
content-type · 11 sites
The plugin's template code fetches arbitrary public web content (via REST API, HTML scraping, RSS parsing, and Playwright) and embeds it directly into LLM prompts via _build_prompt(), creating an indirect prompt injection vector where untrusted third-party content influences AI scoring and filtering decisions.
news.ycombinator.com
domain · 1 site
Hacker News is explicitly listed as a real-world scraping target whose user-generated content would be fetched and fed into LLM prompts for analysis.
SKILL.md
748
"Build me an agent that monitors Hacker News for AI startup funding news"
linkedin.com
domain · 1 site
LinkedIn is explicitly listed as a real-world scraping target whose job listing content would be fetched and fed into LLM prompts for scoring.
SKILL.md
751
"Collect Chief of Staff job listings from LinkedIn and Cutshort into Notion"
cutshort.io
domain · 1 site
Cutshort is explicitly listed as a real-world scraping target whose job listing content would be fetched and fed into LLM prompts for scoring.
SKILL.md
751
"Collect Chief of Staff job listings from LinkedIn and Cutshort into Notion"
reddit.com
domain · 1 site
Reddit subreddits are explicitly listed as a real-world scraping target whose user-generated posts would be fetched and fed into LLM prompts for sentiment classification.
SKILL.md
752
"Monitor a subreddit for posts mentioning my company — classify sentiment"
arxiv.org
domain · 1 site
arXiv is explicitly listed as a real-world scraping target whose academic paper content would be fetched and fed into LLM prompts for analysis.
SKILL.md
753
"Scrape new academic papers from arXiv on a topic I care about daily"
github.com
domain · 2 sites
GitHub is explicitly listed as a real-world scraping target whose repository content would be fetched and fed into LLM prompts for summarization.
SKILL.md
90
- GitHub repos → summarise new releases
SKILL.md
750
"Track new GitHub repos tagged with 'llm' or 'agents' — summarise each one"
- Audited
- Security analysis