slack

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

• Third-party content exposure detected (high risk: 0.85). SKILL.md describes using browser automation to take Slack “snapshot”/“get text” from the currently logged-in Slack UI, which will include outsider-authored message text (other users’ chat content) that can be ingested into the agent’s LLM context via extracted text/snapshots at runtime.