CVE-2026-65699
MEDIUM
4.2
CVSS 3.1
Description
AgentGPT through 1.0.0 contains an authorization bypass through user-controlled key vulnerability that allows authenticated users to attach tasks to another user's agent run by supplying a target run_id in the request body without ownership verification. The AgentCRUD.create_task and validate_task_count functions look up the target AgentRun using the client-supplied run_id without confirming the run belongs to the requesting user, enabling an attacker who obtains a valid run_id to corrupt task history, exhaust the per-run loop budget, and drive LLM costs against the victim's run.
Metadata
Severity & Metrics
4.2
MEDIUM CVSS 3.1
CVSS:3.1/AV:N/AC:H/PR:L/UI:N/S:U/C:N/I:L/A:L
SSVC — CISA Coordinator
Affected products (1)
| Vendor | Product | Platform | Versions |
|---|---|---|---|
| reworkd | AgentGPT | — | 0 ≤ 1.0.0 |
Weakness (CWE)
| CWE | Source | Description |
|---|---|---|
| CWE-639 | cna | Authorization Bypass Through User-Controlled Key |
CVSS scores (2)
| Score | Severity | Version | Source | Vector |
|---|---|---|---|---|
| 4.2 | MEDIUM | 3.1 | cna | CVSS:3.1/AV:N/AC:H/PR:L/UI:N/S:U/C:N/I:L/A:L |
| 2.3 | LOW | 4.0 | cna | CVSS:4.0/AV:N/AC:H/AT:N/PR:L/UI:N/VC:N/VI:L/VA:L/SC:N/SI:N/SA:N |
References (2)