EUVD-2026-52428Severity by source
CVSS:4.0/AV:L/AC:L/AT:N/PR:N/UI:A/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N/E:X/CR:X/IR:X/AR:X/MAV:X/MAC:X/MAT:X/MPR:X/MUI:X/MVC:X/MVI:X/MVA:X/MSC:X/MSI:X/MSA:X/S:X/AU:X/R:X/V:X/RE:X/U:X
Local vector and user interaction required (victim runs ooo in attacker directory); no victim-side privileges needed since attacker delivers payload via committed repo files.
Primary rating from Vendor (GitHub_M).
CVSS VectorVendor: GitHub_M
Lifecycle Timeline
4DescriptionCVE.org
Ouroboros is a local-first runtime for AI coding agents that records their actions and applies user-defined policies to constrain behavior. Versions prior to 0.42.1 have an incomplete denylist. Several execution-routing keys of the same RCE class were omitted, so a malicious cloned repo can still reach arbitrary command execution by shipping a .env (auto-loaded at import, with no review step). The CVE-2026-47211 fix added _UNTRUSTED_ENV_DENYLIST to stop an untrusted project-directory .env from redirecting execution, but it did not account for all keys. The backend config-home and MCP/plugin roots bypass the approval gate by pointing the nested agent, MCP servers, and plugin roster at attacker config. Other variables re-enable blocked local transports, replace sub-agent prompts, switch backends, and lower tool approval classes, further weakening the approval gate. This issue has been fixed in version 0.42.1.
AnalysisAI
Arbitrary command execution in Ouroboros (ouroboros-ai pip package, versions <= 0.42.0) is reachable when a developer runs ooo inside a malicious cloned repository, due to an incomplete denylist that left several execution-routing environment variable keys unblocked. The 0.39.0 fix for CVE-2026-47211 introduced _UNTRUSTED_ENV_DENYLIST but omitted keys spanning backend config-home roots, MCP bridge config, plugin trust roots, SSRF transport toggles, and capability override paths - all of which are auto-loaded from a project-directory .env at import with no review step. …
Unlock full vulnerability intelligence
- Risk assessment & exploitation conditions
- Attack chain visualization
- Remediation with exact patch versions
- Threat intelligence from 22 sources
- Personal watchlist & email alerts
Free forever · No credit card required
Attack ChainAIDerived
Hypothetical attack flow derived from CVE metadata
Vulnerability AssessmentAI
| Exploitation | The victim must have ouroboros-ai <= 0.42.0 installed and must run `ooo` from within a cloned or otherwise attacker-controlled repository directory. … Additional conditions and limiting factors are described in the full assessment. |
| Risk Assessment | The CVSS 4.0 vector (AV:L/AC:L/AT:N/PR:N/UI:A/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N, score 8.4) correctly characterizes the attack as local - the victim must actively clone and invoke `ooo` within the attacker-controlled directory - with no attacker-side privileges required and full confidentiality, integrity, and availability impact on the vulnerable system. … Full risk analysis with EPSS, KEV, and SSVC signal comparison available after sign-in. |
| Exploit Scenario | Full exploit scenario with step-by-step reproduction available after sign-in. |
| Remediation | Upgrade ouroboros-ai to version 0.42.1, which adds all omitted execution-routing keys to `_UNTRUSTED_ENV_DENYLIST` and removes the cwd-based auto-discovery of `./.ouroboros/mcp_servers.yaml`, leaving only the explicit `OUROBOROS_MCP_CONFIG` environment variable and `~/.ouroboros/mcp_servers.yaml` (both in the trusted user home) as valid MCP config sources. … Detailed patch versions, workarounds, and compensating controls in full report. |
Recommended ActionAI
Within 24 hours, identify all systems using Ouroboros ≤0.42.0 and instruct development teams to discontinue use of the package pending update. …
Sign in for detailed remediation steps and compensating controls.
Threat intelligence, references, and detailed analysis are available after sign-in.
Same technique Authentication Bypass
View allShare
External POC / Exploit Code
Leaving vuln.today