Attack ChainAIDerived
Hypothetical attack flow derived from CVE metadata – not a confirmed exploit chain
-
AccessReach backend over network
-
DeliverySend GET /v1/workspaces unauthenticated
-
ExploitReceive full workspace list
-
ExecutionRead unmasked browserfabric_api_key and creator emails
-
ImpactAbuse leaked API keys downstream
Vulnerability Assessment
| Weakness |
CWE
OWASP Top 10:2025
A07:2025 - Authentication Failures
ASVS 5.0
V6 - Authentication Authentication required for critical functions
MITRE ATT&CK
1 technique across 1 tactic: Initial Access The technique list with ATT&CK links and the kill-chain view is part of the Basic plan and up. |
| Exploitation Conditions | No special conditions - remote unauthenticated exploitation against any network-reachable OpenAgents Workspace backend through launcher-v1.0.17, by sending a single GET /v1/workspaces request (the endpoint performs no authentication per CWE-306). The only limiting factor is network reachability of the backend: exploitation requires the attacker to reach the HTTP API, so instances bound to localhost, placed behind an authenticating reverse proxy, or otherwise network-segmented are not directly exposed. Severity is amplified where the browserfabric_api_key is populated, since that secret is returned unmasked. |
| Impact | Multiple signals align on genuine confidentiality risk. The CVSS 4.0 vector (AV:N/AC:L/AT:N/PR:N/UI:N/VC:H) confirms a network-reachable, low-complexity flaw requiring no authentication (PR:N) and no user interaction - fully consistent with the CWE-306 (Missing Authentication) classification and the description's unauthenticated GET /v1/workspaces endpoint. The 8.7 score is driven entirely by high confidentiality impact (VC:H) with no integrity or availability impact (VI:N/VA:N), which matches an information-disclosure-only bug. No EPSS score and no KEV status were provided in the input, so exploitation probability and active-exploitation status cannot be confirmed from available data; absence of these signals is not evidence of low risk. A VulnCheck advisory plus a public HackMD technical write-up exist, but no explicit POC flag was supplied - treat exploit availability as a technical write-up rather than confirmed weaponized code. The real-world priority is tempered by the product's likely small deployment footprint: this is a high-severity flaw in a niche open-source project, so it is a real must-fix for any organization running OpenAgents Workspace but not a broad fleet-wide emergency. The most severe consequence is leakage of the unmasked browserfabric_api_key, which enables downstream credential abuse against the BrowserFabric service - that secondary blast radius should be weighed above the directory-listing of emails/slugs. |
| Exploit ScenarioDerived | An unauthenticated attacker who can reach the OpenAgents Workspace backend sends a single GET request to /v1/workspaces and receives a JSON listing of every workspace, including each workspace's settings map with the unmasked browserfabric_api_key, plus ids, slugs, creator emails, and member lists. The attacker then reuses the harvested API keys against the BrowserFabric service and uses the email/member data for targeted follow-on attacks. A public HackMD technical write-up documents the flaw, though no confirmed standalone exploit tool was indicated in the input. |
Get an e-mail when this CVE is exploited, patched or re-scored. Free account.
Alert meAnalysisAIv1 – Oct 11, 2026
Unauthenticated information disclosure in OpenAgents Workspace (through launcher-v1.0.17) enables remote attackers to list all workspaces and retrieve sensitive data-including the unmasked browserfabric_api_key, workspace IDs, slugs, creator emails, and member lists-via a single GET /v1/workspaces request. The endpoint lacks authentication (CWE-306), so exploitation requires only network reachability to the backend API. No public exploit code or CISA KEV listing is reported; however, the exposure of API credentials creates a high confidentiality risk, consistent with the CVSS 4.0 base score of 8.7 and independent assessment.
RemediationAI
No vendor-released patch version was identified in the available data. Until a fix is available, immediately restrict network access to the OpenAgents Workspace backend: bind the service to localhost or place it behind an authenticating reverse proxy, and use firewall rules to block external access to the /v1/workspaces endpoint. These controls prevent unauthenticated exploitation but may disrupt legitimate remote API consumers. As a code-level mitigation, add authentication (e.g., API keys or OAuth) to the /v1/workspaces endpoint and mask or remove the browserfabric_api_key from the settings map returned in responses; this requires modifying the backend source as indicated in the GitHub references. Rotate any browserfabric_api_key that may have been exposed, accepting that existing integrations using the old key will break until updated. Monitor access logs for unauthorized GET /v1/workspaces requests. Refer to the VulnCheck advisory (https://www.vulncheck.com/advisories/openagents-workspace-through-launcher-1.0.17-unauthenticated-credential-exposure-via-v1-workspaces) for updates and the HackMD write-up (https://hackmd.io/@haind/openagents-workspace-list-browserfabric-key-disclosure) for technical details.
Affected VersionsENISA
Affects your stack
Find out whether the stacks and servers you run are on an affected version.
Checking this CVE against your own stacks and servers is part of the Basic plan and up.
Technical ContextAI · show details
OpenAgents is a platform for AI agents, and its Workspace backend is a Python-based HTTP API that manages workspaces. The vulnerability resides in the /v1/workspaces endpoint, as shown in the GitHub source references at workspaces.py lines 135-177 and 326-349, which handle listing workspaces and fetching workspace details without authentication checks. This is a classic CWE-306 (Missing Authentication for Critical Function) flaw: the endpoint is reachable without credentials and returns workspace objects that include a settings map with the browserfabric_api_key in plaintext (unmasked), along with workspace metadata like IDs, slugs, creator emails, and member lists. The CPE cpe:2.3:a:openagents-org:openagents:*:*:*:*:*:*:*:* indicates all versions of the openagents-org software are potentially affected, but the specific vulnerable component is the Workspace backend through launcher-v1.0.17. The CVSS vector (CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:H/VI:N/VA:N/SC:N/SI:N/SA:N) confirms remote, unauthenticated access with high confidentiality impact only.
Know when Openagents gets its next vulnerability
- Watch products and get an email when something new is published
- Check this CVE against your own stack and servers
- Exploit and PoC references, threat intelligence from 22 sources
No credit card · 7-day full trial
Severity by source
CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:N/VC:H/VI:N/VA:N/SC:N/SI:N/SA:N
Unauthenticated network endpoint (AV:N/PR:N) trivially queried (AC:L/UI:N) leaking secrets gives C:H; no integrity or availability impact and no scope change.
Primary rating from Vendor (VulnCheck).
CVSS VectorVendor: VulnCheck
DescriptionCVE.org
OpenAgents Workspace backend through launcher-v1.0.17 contains an information disclosure vulnerability that allows unauthenticated attackers to list all workspaces via GET /v1/workspaces. Attackers can read the unmasked browserfabric_api_key in each workspace's settings map, along with workspace ids, slugs, creator emails and member lists.
Executive ViewAI
An information-disclosure flaw in OpenAgents Workspace (through launcher-v1.0.17) allows anyone reaching the backend API to pull the complete workspace listing along with unmasked credentials, including the browserfabric_api_key, plus workspace IDs, slugs, creator emails, and member lists. Because a single request returns live API keys alongside organizational and personnel data, the exposure can cascade into unauthorized access to connected services, making this an enterprise-wide confidentiality issue rather than a narrow technical bug. No vendor patch is currently available, so leadership should assume the affected data is retrievable until network access is contained.
No vendor-released patch identified at time of analysis. Interim mitigations: restrict network reachability to the backend API (VPN, IP allowlist, or authenticated reverse proxy); block external access to the GET /v1/workspaces endpoint; rotate the browserfabric_api_key and any other credentials exposed through that response; review and alert on access logs for enumeration of workspace listings; and monitor the vendor advisory channel for a launcher-v1.0.17 fix.
Threat Intelligence Overview
How this decision was reached
The engine reads each signal below and applies a fixed list of rules from top to bottom. The first rule that matches sets the recommended action – like a statement, every line is shown.
What every action level means
SSVC AssessmentENISA
Priority Score
Attack Technique
Authentication bypass attacks exploit flaws in the verification mechanisms that control access to systems and applications.
Information disclosure occurs when an application unintentionally exposes sensitive data that aids attackers in reconnaissance or directly compromises security.
More in Openagents
View allServer-side request forgery in openagents up to 0.8.19 and 0.9.3.post20 allows unauthenticated remote attackers to manip
Path traversal vulnerability in xlang-ai OpenAgents up to commit ff2e464 allows attackers to read or write arbitrary fil
EUVD-2026-96905
GHSA-66fr-h599-fx6p
NVD
CVE.org
EPSS
Exploit-DB