amz-listing-suppression-recovery

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.85). The skill’s runtime workflow ingests “Amazon’s exact wording” (suppression text) and uses it to generate output; this is outsider-authored free text from Amazon’s system, which the agent reads and places into the LLM context.

HIGH W008: Secret detected in skill content (API keys, tokens, passwords).

• Secret detected (high risk: 1.00). I scanned the skill text for literal, high-entropy credentials. The only string resembling an access token is the WhatsApp invite URL:

https://chat.whatsapp.com/ILX65p1yWcaIG3c9WGHpTY

The trailing token (ILX65p1yWcaIG3c9WGHpTY) is a non-placeholder, high-entropy join token that grants access to the WhatsApp group — it functions as a usable credential. No API keys, private-key blocks, or other high-entropy secrets were found. Other values in the doc are plain text, placeholders, or low-entropy examples and were ignored.