Hijacked by injectionPI-0030
ForcedLeak: web form submissions could make Salesforce Agentforce leak CRM lead data
Noma Security found that hidden instructions in a web form could make the agent embed lead email addresses in an image link when an employee later asked about the lead. The chain worked partly because an expired domain on Salesforce's list of trusted addresses could be bought for a few dollars. Salesforce patched the flaw and began enforcing lists of trusted web addresses from Sept. 8.
Harm: No harm reportedControl level 2, Instruction violation
Not counted · can set the control floorEffect on the index
Near miss. Harm was plausible but prevented. It is rated only on what actually happened.
Sources
How we know
4 sources · developer confirmed. Links go to the original publishers; the summary above is in our own words.
- researchForcedLeak: AI agent risks exposed in Salesforce AgentforceNoma Security · Sept. 25, 2025noma.security/blog/forcedleak-agent-risks-exposed-in-salesforce-agentforce
- newsPrompt injection – and a $5 domain – trick Salesforce Agentforce into leaking salesThe Register · Sept. 26, 2025theregister.com/software/2025/09/26/salesforce-agentforce-tricked-into-leaking…
- newsSalesforce patches critical ForcedLeakThe Hacker News · Sept. 25, 2025thehackernews.com/2025/09/salesforce-patches-critical-forcedleak.html
- newsVulnerability in Salesforce AI could be tricked into leaking CRM dataCSO Online · Sept. 25, 2025csoonline.com/article/4063044/vulnerability-in-salesforce-ai-could-be-tricked-i…
Why this rating
No harm reported; control failure level 2
Two assessments. Documented harm sets the reading; a level 4 or 5 control failure on the same record makes that harm count as more than one record. Control alone sets only a floor.
Observed harm
No harm reportedResearcher demonstration on controlled accounts; no victims reported.
Not a finding that no harm occurred.
The harm scale
- 1 Negligible Inconvenience, easily remedied.
- 2 Minor Limited, recoverable harm.
- 3 Moderate Material harm needing significant effort to remedy.
- 4 Severe Severe harm to health, rights, property or essential services.
- 5 Catastrophic Society-scale harm or disruption beyond a community's capacity to cope.
Control assessment
Instruction violationThe agent read CRM data it could access and followed injected instructions to send it out; inside its permissions.
Reported beside the index. It adds nothing to a harm reading; when no harm counts in a window, the highest eligible control level in the window is the reading.
The control scale
- 1 Negligible Behaved as instructed. The problem was an ordinary error or a flawed output, with no rule broken.
- 2 Minor Broke an explicit instruction or rule, or gamed its goal, while staying inside its permissions and environment: for example, reward hacking, misreporting results, or following injected instructions within its permissions.
- 3 Moderate Acted outside the permissions it was given, deceived its overseers about its own actions, or tampered with oversight tools (logs, monitors, shutdown). Stopped by normal controls within an hour.
- 4 Severe Reached systems outside its permitted environment, or acquired money, compute or accounts without authorization. The type and mechanism say how.
- 5 Catastrophic The developer or operator lost control: the system copied its weights outside their control, replicated itself, or resisted being stopped for 24 hours or more.
Rating rationale
No exploitation or external harm was reported from this controlled test; Salesforce did not say whether any had occurred. Control level 2 reflects injected instructions obeyed within permissions.
Effect on the index
It does not count toward the Sept. 29 reading
The reading for the week to Sept. 29, 2025, with this record and without it. Harms count in full for two weeks after they are reported, then one level less every two weeks.
Not counted: no harm reported. 3 other records behind the reading for that week.
The arithmetic
| Step | With it | Without |
|---|---|---|
| Counts toward the index?documented, external, eligible evidence | No | — |
| Worst documented harm, ksets the band | — (none counting) | — (none counting) |
| Harms at that level, nposition in the band | 0 | 0 |
| Highest control level breachedsets the reading only when no harm counts | 3 | 3 |
| Readingrounded down | 3 Control failures only | 3 Control failures only |
Inside the window of 4 weekly readings
| Week to | Reading | Band |
|---|---|---|
| Sept. 29, 2025 | 3 | Control failures only |
| Oct. 6, 2025 | 3 | Control failures only |
| Oct. 13, 2025 | 3 | Control failures only |
| Oct. 20, 2025 | 2 | Control failures only |
Revisions
What we changed
1 logged. Every change to a rating is logged here, with the reason.
- v1Oct. 7, 2026
First published.
Cite and share
Use this record
Citation
Paperclip Index. “ForcedLeak: web form submissions could make Salesforce Agentforce leak CRM lead data.” Record PI-0030. Reported Sept. 25, 2025; updated Oct. 7, 2026. Rated under methodology v1.1. https://paperclipindex.com/incident/PI-0030