CVE-2026-77938highCWE-526

CVE-2026-77938: high-severity vulnerability in letta-ai letta

Letta through 0.16.8 Credential Exposure via Tool Sandbox Environment Inheritance

Published

Follow this CVE+ Monitor letta

E-mail alert when it lands in KEV, gets a public PoC or its score goes up.

41Vexday Risk Score

No sign of exploitation. It has a public proof of concept.

ssvc Attendcvss 7.1epss 0.4%
Letta through 0.16.8 contains an information disclosure vulnerability that allows users running tools to read server secrets because LocalSandbox subprocesses inherit the full server environment. Attackers can run code via POST /v1/tools/run or custom tools to read LETTA_PG_URI, POSTGRES_PASSWORD, and provider API keys, then access cross-tenant database data.
CVSS:4.0/AV:N/AC:L/AT:N/PR:L/UI:N/VC:H/VI:N/VA:N/SC:N/SI:N/SA:N
Affected products
letta-ai · letta
⚠ Public resources, to assess the exposure of systems you control or are authorized to test. Test only with authorization.
Related CVEs — letta-ai letta

In the same product, most dangerous first.

CVE-2026-4965MEDIUMletta-ai letta Incomplete Fix CVE-2025-6101 ast_parsers.py resolve_type eval injectionEPSS 0.6%CVE-2026-77936CRITICALLetta through 0.16.8 Insecure Default Configuration Unauthenticated API and PostgreSQL AccessEPSS 0.6%CVE-2026-77937HIGHLetta through 0.16.8 Cross-Tenant Impersonation via user_id HeaderEPSS 0.5%CVE-2026-4964MEDIUMletta-ai letta File URL message_helper.py _convert_message_create_to_message server-side request forgeryEPSS 0.3%CVE-2025-6101MEDIUMletta-ai letta interface.py function_message eval injectionEPSS 0.3%