← Back to search

CVE-2026-123786

8.5 HIGHpublic exploit available

Published 2026-11-06 · Updated 2026-11-09

AI risk analysis

Summary
Agenta before 0.122.5 has a server-side request forgery (SSRF) flaw where users with workflow edit permissions can set arbitrary revision URLs, causing the server to make unintended requests. This allows attackers to exfiltrate the secret-resolve Authorization token and access sensitive provider secrets or internal/cloud metadata endpoints.
Exploitability
Exploitation requires low skill and only a workspace membership with workflow edit rights; no user interaction or external tools are needed beyond standard HTTP requests.
Blast radius
Successful exploitation enables full compromise of cloud provider secrets stored in the vault, potential access to internal infrastructure via metadata endpoints, and limited integrity impact through forced server-side actions.
Detection
Defenders should monitor for anomalous outbound HTTP requests from the Agenta application to unexpected external domains or internal IP ranges (e.g., 169.254.169.254), especially those including Authorization headers containing secret-resolve tokens; log entries showing workflow invocations with non-standard revision URLs pointing to untrusted domains may also indicate exploitation attempts.
Prioritized remediation
Upgrade to Agenta version 0.122.5 or later.
ssrfsecrets-exposureauth-bypassweb-appcloud-metadata

Analysis generated locally by garnet-coder (no data left the box). AI-assisted — verify against primary sources before acting.

NVD description

Agenta before 0.122.5 contains a server-side request forgery vulnerability that allows workspace members with workflow edit rights to redirect workflow invocations by setting arbitrary revision URLs. Attackers can capture the attached secret-resolve Authorization token to read provider secrets from the vault API, or reach internal and cloud metadata addresses.

CVSS vector

CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:C/C:H/I:L/A:N

Weaknesses

CWE-918

Public exploit & PoC references

All references

Source data: NVD (nvd.nist.gov), public domain. Exploit-DB.ai adds local AI analysis for defensive use only.

Related CVEs

Related by shared AI tags and CWE weakness class. Browse the full archive.