GoClaw CVE-2026-16199

Exploitation requires a valid low-privileged authenticated account on the target GoClaw instance (consistent with PR:L in the CVSS 4.0 vector). The vulnerable code path is the ExecTool.Execute function in goclaw/internal/tools/credentialed_exec.go, meaning the service must be network-reachable and the attacker must be able to invoke this function. No additional special configuration or non-default deployment mode is indicated as required beyond having credentials. Limiting factors: unauthenticated exploitation is not supported (PR:L rules out PR:N scenarios); the product must be actively deployed and the ExecTool functionality must be in use; impact is bounded to partial (L) confidentiality, integrity, and availability - full system compromise is not indicated by available data.

The CVSS 4.0 score of 2.1 (vector AV:N/AC:L/AT:N/PR:L/UI:N/VC:L/VI:L/VA:L/SC:N/SI:N/SA:N/E:P) reflects low real-world severity. While the attack vector is network and complexity is low, exploitation requires a low-privileged authenticated account (PR:L), limiting opportunistic mass exploitation. The E:P modifier confirms a publicly available proof-of-concept, which elevates the likelihood of targeted exploitation in environments where GoClaw is deployed. Impact metrics (VC:L/VI:L/VA:L) indicate only partial compromise of the vulnerable system with no scope change to subsequent systems (SC/SI/SA all N). No KEV listing is present, and no EPSS score was provided in the input data. The combination of a low-privilege prerequisite, limited impact ceiling, and niche product footprint makes this a low-priority item for most organizations, though environments actively using GoClaw should remediate given the public PoC. There is no CVSS 3.x vector in the input - the 3.1 assessment below is independently inferred.

An attacker with a valid low-privileged account on a GoClaw instance sends a crafted request to the ExecTool.Execute function, supplying manipulated parameters that bypass the inadequate authorization check. Because a public proof-of-concept exists in GitHub issue #1215, the mechanics of the manipulation are already publicly documented, lowering the bar for exploitation by less skilled attackers in targeted engagements against known GoClaw deployments.