jamesmoss/random-lunch
Orders a random lunch for me from Feedr.co.
43
54%
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
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
SKILL.md explicitly instructs the agent to use a Chrome plugin to visit the public site https://feedr.co/en-gb/canteen and read/interpret its pages to find days and choose/place orders, so untrusted third-party web content could influence actions.
Where we found it
feedr.co
domain · 2 sites
The plugin instructs the agent to visit feedr.co in a browser, read its page content (available days, dishes, prices), and take actions (select and order dishes) based on that untrusted third-party web content.
SKILL.md
6
Use the Chrome plugin to visit https://feedr.co/en-gb/canteen and find any days
SKILL.md
3
description: Order's a random lunch for me from Feedr.co.
- Audited
- Security analysis