Incident · 2025 · Prompt injection

Salesforce Agentforce

The ForcedLeak research showed CRM data leaving through an image link to a domain that was still on the allow-list after it had expired. The researchers bought it for $5.

What happened

The ForcedLeak research showed CRM data leaving through an image link to a domain that was still on the allow-list after it had expired. The researchers bought it for $5.

Source: Noma Security ↗. Follow it for the full account.

The cause: prompt injection

Text the assistant reads (an email, a web page, a file name) carries instructions, and the assistant follows them as if they came from you.

Framework entries that cover it: OWASP LLM01: Prompt Injection · MITRE ATLAS: LLM Prompt Injection

The solution
  1. 1

    What would have caught this one

    Exfiltration tests through links, images and every other channel the assistant can render.

  2. 2

    Prevent the whole class

    Treat everything the assistant reads as untrusted. Separate reading from acting, and require confirmation before any action that sends data out.

  3. 3

    Test for it

    Plant instructions in every field the assistant can read, including encoded and split ones, and check whether any action or data leaves.