CVE-2026-54142
OpenHuman: Indirect prompt injection via background email sync executes Composio actions without user approval
OpenHuman is an open-source agent harness available as a desktop application and Rust library. Before 0.57.13, background synchronization for connected services such as email, Slack, and Notion stores raw external content and later places it in the model context of autonomous background agents without neutralizing embedded instructions. ApprovalGate treats tool calls without an interactive chat context as pre-authorized automation, so attacker-authored content processed during a background tick can influence the model to invoke Composio actions without presenting an approval request. An attacker who can deliver content to a connected account can use this path to send or forward data, post messages, create or modify third-party content, or disclose information from OpenHuman memory and integrations. The issue requires an active integration and successful model interpretation of the injected instructions. This issue is fixed in version 0.57.13.
Attack profile
How easy it is to exploit
- Attack vector
- Network
- Complexity
- High
- Privileges needed
- None
- User interaction
- None
What an attacker can affect
- Confidentiality
- High
- Integrity
- High
- Availability
- None
Exploit likelihood
What each source reports
CVE List Score used
8.7
OpenHuman: Indirect prompt injection via background email sync executes Composio actions without user approval
OpenHuman is an open-source agent harness available as a desktop application and Rust library. Before 0.57.13, background synchronization for connected services such as email, Slack, and Notion stores raw external content and later places it in the model context of autonomous background agents without neutralizing embedded instructions. ApprovalGate treats tool calls without an interactive chat context as pre-authorized automation, so attacker-authored content processed during a background tick can influence the model to invoke Composio actions without presenting an approval request. An attacker who can deliver content to a connected account can use this path to send or forward data, post messages, create or modify third-party content, or disclose information from OpenHuman memory and integrations. The issue requires an active integration and successful model interpretation of the injected instructions. This issue is fixed in version 0.57.13.
- Record
CVE-2026-54142- CVSS 3.1
- 8.7 High
CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:C/C:H/I:H/A:N - Weakness
- CWE-1427
- Updated
- 2026-11-05
EUVDEUVD-2026-117597
8.7
OpenHuman is an open-source agent harness available as a desktop application and Rust library. Before 0.57.13, background synchronization for connected services such as email, Slack, and Notion stores raw external content and later places it in the model context of autonomous background agents without neutralizing embedded instructions. ApprovalGate treats tool calls without an interactive chat context as pre-authorized automation, so attacker-authored content processed during a background tick can influence the model to invoke Composio actions without presenting an approval request. An attacker who can deliver content to a connected account can use this path to send or forward data, post messages, create or modify third-party content, or disclose information from OpenHuman memory and integrations. The issue requires an active integration and successful model interpretation of the injected instructions. This issue is fixed in version 0.57.13.
- Record
EUVD-2026-117597- CVSS 3.1
- 8.7 High
CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:C/C:H/I:H/A:N - Updated
- 2026-11-05
NVD
8.7
OpenHuman is an open-source agent harness available as a desktop application and Rust library. Before 0.57.13, background synchronization for connected services such as email, Slack, and Notion stores raw external content and later places it in the model context of autonomous background agents without neutralizing embedded instructions. ApprovalGate treats tool calls without an interactive chat context as pre-authorized automation, so attacker-authored content processed during a background tick can influence the model to invoke Composio actions without presenting an approval request. An attacker who can deliver content to a connected account can use this path to send or forward data, post messages, create or modify third-party content, or disclose information from OpenHuman memory and integrations. The issue requires an active integration and successful model interpretation of the injected instructions. This issue is fixed in version 0.57.13.
- Record
CVE-2026-54142- CVSS 3.1
- 8.7 High
CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:C/C:H/I:H/A:N - Updated
- 2026-11-07
OSV
8.7
OpenHuman: Indirect prompt injection via background email sync executes Composio actions without user approval
OpenHuman is an open-source agent harness available as a desktop application and Rust library. Before 0.57.13, background synchronization for connected services such as email, Slack, and Notion stores raw external content and later places it in the model context of autonomous background agents without neutralizing embedded instructions. ApprovalGate treats tool calls without an interactive chat context as pre-authorized automation, so attacker-authored content processed during a background tick can influence the model to invoke Composio actions without presenting an approval request. An attacker who can deliver content to a connected account can use this path to send or forward data, post messages, create or modify third-party content, or disclose information from OpenHuman memory and integrations. The issue requires an active integration and successful model interpretation of the injected instructions. This issue is fixed in version 0.57.13.
- Record
CVE-2026-54142- CVSS 3.1
- 8.7 High
CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:C/C:H/I:H/A:N - Weakness
- CWE-1427
- Updated
- 2026-11-07
Affected products
History
| Recorded | Changed | Sources |
|---|---|---|
| aliases, references | NVD, OSV | |
| references | CVE List, EUVD, NVD | |
| aliases, published | EUVD | |
| first indexed | CVE List |
Dates are when Every CVE recorded each change. Upstream dates may differ.
In short
CVE-2026-54142 is a high vulnerability (CVSS 8.7) in Tinyhumansai Openhuman, published November 5, 2026. No source lists public exploit code. EPSS puts the chance of exploitation in the next 30 days at 0.52%, the 44th percentile. No source names a fixed version.
References
github.com/tinyhumansai/openhuman/security/advisories/GHSA-p9gw-q7cf-57qr
github.com/tinyhumansai/openhuman/pull/3227
github.com/tinyhumansai/openhuman/commit/3a7180da7765af861bf44d79e025fc67dbeae0ca
github.com/tinyhumansai/openhuman/releases/tag/v0.57.13
github.com/CVEProject/cvelistV5/tree/main/cves/2026/54xxx/CVE-2026-54142.json
nvd.nist.gov/vuln/detail/CVE-2026-54142