amz-ip-complaint-retraction-kit

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

• Third-party content exposure detected (high risk: 0.75). The skill’s runtime workflow ingests the complaint notification text (including rights owner name/email, complaint ID, ASIN, and complaint details) and may also fetch rights-owner contact info from USPTO TESS / US Copyright Office; both are outsider-authored free text that can be included in the agent’s LLM context via the complaint/lookup content.

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

• Secret detected (high risk: 1.00). I scanned the skill content for high-entropy literal values that provide access.

Findings:

• The WhatsApp invite URL ("https://chat.whatsapp.com/ILX65p1yWcaIG3c9WGHpTY") contains a high-entropy invite token that grants access to a private group — this qualifies as a secret/credential under the definition (literal, random-looking value that provides access).
• Other values in the document are not secrets: the example numeric Complaint ID is just an example number, the Amazon support email (notice-dispute@amazon.com) is a public contact address, and names/descriptions are documentation content. No API keys, private keys, or passwords were found.

Therefore I treat the WhatsApp invite link as an active secret present in the document.