CVE-2026-108756: Missing Authorization in Abilityai trinity
Description
Abilityai Trinity versions up to 0.9.5 have a missing authorization vulnerability in the Telegram router component. This flaw allows agent-scoped MCP API keys to perform operations intended only for human users, such as sending messages via the owner's bot token, replacing the binding with another token, or deleting it. The vulnerability can be exploited by attackers controlling an agent, often through prompt injection techniques.
CVSS v4.0
Score 5.3medium
Affected software
Abilityai
trinity
Run on your own infrastructure? Check whether these packages are installed with threat-finder — our free open-source scanner.
AI-Powered Analysis
Machine-generated threat intelligence
Technical Analysis
CVE-2026-108756 describes a missing authorization vulnerability in Abilityai Trinity through version 0.9.5. The issue resides in the Telegram router, where agent-scoped MCP API keys are improperly authorized to perform human-only binding operations. Attackers who gain control over an agent—commonly via prompt injection—can misuse the owner's bot token to send messages, replace the binding with their own token, or delete the binding entirely. This vulnerability has a CVSS 4.0 base score of 5.3, indicating a medium severity level.
Potential Impact
Exploitation of this vulnerability allows an attacker controlling an agent to send messages as the bot owner, replace the bot token binding with their own token, or delete the binding. This could lead to unauthorized message sending and potential disruption or takeover of bot operations. However, the vulnerability requires agent control and does not allow privilege escalation beyond the agent scope.
Mitigation Recommendations
Patch status is not yet confirmed — check the vendor advisory for current remediation guidance. Since no patch or official fix information is provided, users should monitor vendor communications for updates. Until a fix is available, restrict agent control and avoid exposure to prompt injection attacks that could lead to agent compromise.
Technical Details
- Data Version
- 5.2
- Assigner Short Name
- VulnCheck
- Date Reserved
- 2026-10-11T01:57:24.159Z
- Cvss Version
- 4.0
- State
- PUBLISHED
Threat ID: 6acb85c42cdf04f656d09a34
Added to database: 10/11/2026, 12:49:08 UTC
Last enriched: 10/11/2026, 13:03:39 UTC
Last updated: 10/11/2026, 21:33:13 UTC
Views: 19
Community Reviews
0 reviewsCrowdsource mitigation strategies, share intel context, and vote on the most helpful responses. Sign in to add your voice and help keep defenders ahead.
Want to contribute mitigation steps or threat intel context? Sign in or create an account to join the community discussion.
Actions
Updates to AI analysis require Pro Console access. Upgrade inside Console → Billing.
Need more coverage?
Upgrade to Pro Console for AI refresh and higher limits.
For incident response and remediation, OffSeq services can help resolve threats faster.
Latest Threats
Check if your credentials are on the dark web
Instant breach scanning across billions of leaked records. Free tier available.