Description
AgenticMail gives AI agents real email addresses and phone numbers. In @agenticmail/api prior to version 0.9.64, a low-privileged authenticated AgenticMail agent can enumerate another agent's pending/claimed tasks by supplying the target agent name to `GET /api/agenticmail/tasks/pending?assignee=<name>`. The returned task objects include the task IDs and payloads. The same task IDs can then be used with the capability-style task mutation endpoints (`/tasks/:id/claim`, `/tasks/:id/result`, `/tasks/:id/complete`, `/tasks/:id/fail`) to claim, complete, or fail tasks assigned to a different agent. Because ordinary authenticated agents can discover agent names through `GET /api/agenticmail/accounts/directory`, the task ID effectively stops being a secret capability. This turns the intended capability model into a cross-agent authorization bypass. Version 0.9.64 contains a fix.
Published: 2026-07-20
Score: 7.1 High
EPSS: < 1% Very Low
KEV: No
Impact: n/a
Action: n/a
AI Analysis

Impact

AgenticMail allows authenticated agents with low privileges to enumerate the pending tasks of any agent by requesting the pending task list with an arbitrary assignee name. The response includes the identifiers and payload data of each task, which an attacker can then reuse in mutation endpoints to claim, complete, or fail tasks that belong to other agents. This circumvents the intended capability‑based access model; the attacker effectively gains authorization to act on resources they should not control. Inferred from the description, the obtained payload data may contain sensitive information, potentially exposing confidential content to the attacker.

Affected Systems

The vulnerability affects the AgenticMail API service, specifically @agenticmail/api versions older than 0.9.64. Any instance of that API running without the 0.9.64 patch or a later revision is susceptible to this cross‑agent authorization bypass.

Risk and Exploitability

The CVSS v3.1 base score of 7.1 classifies this as a high‑impact vulnerability. The EPSS score of less than 1% indicates a low probability of exploitation in the wild, and the vulnerability is not listed in the CISA KEV catalog. An attacker must be authenticated as a low‑privilege agent and have network access to the API. By enumerating tasks and re‑using their identifiers, the attacker can lift privileges across agents and potentially expose sensitive task content, creating a cross‑agent authorization bypass.

Generated by OpenCVE AI on July 30, 2026 at 18:21 UTC.

Remediation

No vendor fix or workaround currently provided.

OpenCVE Recommended Actions

  • Upgrade AgenticMail API to version 0.9.64 or newer to apply the vendor patch.
  • Configure the API to enforce agent‑specific access controls and deny enumeration of other agents’ tasks for non‑privileged users.
  • Enable audit logging for task enumeration and mutation endpoints, and review logs for anomalous activity to detect potential abuse.

Generated by OpenCVE AI on July 30, 2026 at 18:21 UTC.

Tracking

Sign in to view the affected projects.

Advisories
Source ID Title
Github GHSA Github GHSA GHSA-hjwc-26pj-v3pm AgenticMail: Cross-agent task authorization bypass in AgenticMail API
History

Tue, 28 Jul 2026 21:00:00 +0000

Type Values Removed Values Added
First Time appeared Agenticmail
Agenticmail api
Vendors & Products Agenticmail
Agenticmail api

Tue, 21 Jul 2026 15:30:00 +0000

Type Values Removed Values Added
Metrics ssvc

{'options': {'Automatable': 'no', 'Exploitation': 'poc', 'Technical Impact': 'partial'}, 'version': '2.0.3'}


Mon, 20 Jul 2026 22:15:00 +0000

Type Values Removed Values Added
Description AgenticMail gives AI agents real email addresses and phone numbers. In @agenticmail/api prior to version 0.9.64, a low-privileged authenticated AgenticMail agent can enumerate another agent's pending/claimed tasks by supplying the target agent name to `GET /api/agenticmail/tasks/pending?assignee=<name>`. The returned task objects include the task IDs and payloads. The same task IDs can then be used with the capability-style task mutation endpoints (`/tasks/:id/claim`, `/tasks/:id/result`, `/tasks/:id/complete`, `/tasks/:id/fail`) to claim, complete, or fail tasks assigned to a different agent. Because ordinary authenticated agents can discover agent names through `GET /api/agenticmail/accounts/directory`, the task ID effectively stops being a secret capability. This turns the intended capability model into a cross-agent authorization bypass. Version 0.9.64 contains a fix.
Title AgenticMail: Cross-agent task authorization bypass in AgenticMail API
Weaknesses CWE-639
CWE-862
References
Metrics cvssV4_0

{'score': 7.1, 'vector': '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'}


cve-icon MITRE

Status: PUBLISHED

Assigner: GitHub_M

Published:

Updated: 2026-07-21T14:56:03.184Z

Reserved: 2026-06-24T14:53:40.111Z

Link: CVE-2026-57494

cve-icon Vulnrichment

Updated: 2026-07-21T13:13:44.463Z

cve-icon NVD

No data.

cve-icon Redhat

No data.

cve-icon OpenCVE Enrichment

Updated: 2026-07-30T18:30:05Z

Weaknesses
  • CWE-639

    Authorization Bypass Through User-Controlled Key

  • CWE-862

    Missing Authorization