# CVE-2026-119713: Paperclip remote code execution via execution workspace API

Source: https://junglewise.ai/threats/cve-2026-119713-paperclip-2026-403-0-before-2026-416-0-contains-a-remote-code  
Publisher: Junglewise Threat Intelligence, https://junglewise.ai/threats  
Published: 2026-10-30. Updated: 2026-10-30  
License: CC BY 4.0 with a link back

CVE-2026-119713 · Severity: high · CVSS 8.8 · Published 2026-10-30

Paperclip is an open-source application for managing AI agents in enterprise environments. A vulnerability allows anyone holding an agent API key to inject arbitrary OS commands that execute on the server with the privileges of the Paperclip process when a workspace is archived, potentially giving attackers full control of the server.

Technologies: [Paperclip AI Paperclip](https://junglewise.ai/threats/technologies/paperclip). Vendors: [Paperclip AI](https://junglewise.ai/threats/vendors/paperclip-ai).

## Quick answers

**Is CVE-2026-119713 exploited in the wild?** No. Junglewise Threat Intelligence has no report of CVE-2026-119713 being exploited in the wild.

**How severe is CVE-2026-119713?** High severity, with a CVSS score of 8.8, published 30 October 2026.

**Which products and versions are affected?** Paperclip AI Paperclip 2026.403.0 before 2026.416.0.

**Is there a fix for CVE-2026-119713?** Yes: fixed in 2026.416.0. Junglewise finds every place you run it, drafts the upgrade and applies it once you approve.

**How is CVE-2026-119713 attacked?** It is attacked over the network, in a software library.

**Am I affected by CVE-2026-119713?** Check whether you run Paperclip AI Paperclip at an affected version. Junglewise does that check for you around the clock: connect your code, cloud and domains, and it matches new vulnerabilities like this one against what you actually run and tells you when one reaches you.

## How to fix

- Fixed in 2026.416.0

## Technical details

The vulnerability is a command injection flaw in the execution-workspaces API endpoint. An authenticated attacker with a valid API key can modify the cleanupCommand and teardownCommand fields via PATCH /api/execution-workspaces/:id; these commands are later executed unsanitized through a shell by the cleanupExecutionWorkspaceArtifacts process when the workspace is archived, allowing arbitrary code execution in the security context of the Paperclip server.

## Affected products

- Paperclip AI Paperclip 2026.403.0 before 2026.416.0

## Timeline

- 2026-10-30: disclosed: Vulnerability published

## References

- [https://github.com/paperclipai/paperclip](https://github.com/paperclipai/paperclip)
- [https://github.com/paperclipai/paperclip/blob/v2026.403.0/server/src/routes/execution-workspaces.ts](https://github.com/paperclipai/paperclip/blob/v2026.403.0/server/src/routes/execution-workspaces.ts)

## Related threats

- [paperclip Vulnerable to Unauthenticated Remote Code Execution via Import Authorization Bypass](https://junglewise.ai/threats/cve-2026-41679-paperclip-vulnerable-to-unauthenticated-remote-code-execution-via)
- [Paperclip cross-tenant agent API key IDOR in /agents/:id/keys](https://junglewise.ai/threats/paperclip-idor-in-agent-api-key-management-routes-b978dafc)
- [Paperclip stored XSS in MarkdownBody via javascript: URLs](https://junglewise.ai/threats/paperclip-stored-xss-in-markdownbody-via-javascript-urls-e076e879)
- [Paperclip OS command injection in execution workspace cleanup](https://junglewise.ai/threats/paperclip-os-command-injection-in-execution-workspace-cleanup-1e244286)
- [Paperclip improper access control in codex_local inherits OpenAI Gmail connectors](https://junglewise.ai/threats/paperclip-improper-access-control-in-codex-local-inherits-openai-gmail-185d94bc)