Attack ChainAIDerived
Hypothetical attack flow derived from CVE metadata – not a confirmed exploit chain
-
AccessReach StaffDeck backend over network
-
DeliveryRetrieve hard-coded default app_secret
-
ExploitForge HMAC-SHA256 admin token for tenant_demo
-
ExecutionAuthenticate as tenant administrator
-
ImpactExfiltrate stored model and channel credentials
Vulnerability Assessment
| Weakness |
CWE
OWASP Top 10:2025
A04:2025 - Cryptographic Failures
ASVS 5.0
V11 - Cryptography Key management, no hard-coded keys
|
| Exploitation Conditions | Exploitation requires that the deployment still uses the hard-coded default app_secret shipped in backend/app/config.py (v0.5.6) AND that the seeded 'admin' user within the 'tenant_demo' tenant still exists - both of which are StaffDeck defaults. Given those, an unauthenticated remote attacker can mint a valid HMAC-SHA256 session token (AV:N/PR:N/UI:N). The limiting factors captured by AC:H/AT:P are: operators who rotated/overrode the app_secret or removed the demo tenant/admin account are NOT exploitable, and exploitation is scoped to the 'tenant_demo' tenant's administration (stored model and channel credentials) rather than cross-tenant or host-level compromise. |
| Impact | The CVSS 4.0 base score of 9.2 (Critical) reflects a network-reachable (AV:N), unauthenticated (PR:N), no-user-interaction (UI:N) authentication bypass with full confidentiality, integrity and availability impact to the vulnerable system (VC:H/VI:H/VA:H). Crucially, the vector also carries AC:H and AT:P, signalling that successful exploitation depends on specific preconditions rather than being a trivial drive-by - namely that the hard-coded default app_secret is still in place and the seeded 'tenant_demo'/'admin' account exists. Because the default secret is visible directly in the public source tree (config.py line 10 referenced in the CVE), forging a valid HMAC-SHA256 token is cryptographically trivial for any attacker where defaults were not changed, so the AC:H/AT:P should be read as a configuration-dependency gate, not as genuine attacker difficulty. No EPSS score is provided, so exploitation probability cannot be quantified from the input. The CVE is NOT listed in CISA KEV and no POC flag was supplied, though a detailed public technical writeup (hackmd) describes the signing method. Net assessment: this is a genuinely high-severity flaw for any deployment still running defaults, but its real-world priority is tempered by (a) an obscure, low-footprint product and (b) the hard dependency on an unchanged default secret and seeded demo tenant - treat as high priority for StaffDeck operators specifically, lower as a fleet-wide emergency. Missing signals (no EPSS, no KEV, no explicit POC flag) should be confirmed before final triage. |
| Exploit ScenarioDerived | An attacker who reads the open-source default app_secret from the StaffDeck GitHub repository crafts an HMAC-SHA256 session token for the seeded 'admin' user in 'tenant_demo' and presents it to the backend over the network, gaining full tenant administration without any credentials. From there they read stored model and channel credentials. A detailed public technical writeup describing the signing method exists, though no weaponized exploit code was indicated in the input. |
Get an e-mail when this CVE is exploited, patched or re-scored. Free account.
Alert meAnalysisAIv1 – Oct 16, 2026
Authentication bypass in OpenBMB StaffDeck through 0.5.6 lets unauthenticated attackers forge valid administrator session tokens by signing an HMAC-SHA256 token with the hard-coded default app_secret shipped in backend/app/config.py, targeting the seeded 'admin' user in the 'tenant_demo' tenant. Successful exploitation grants full tenant administration, including access to stored model and channel credentials; the assessed CVSS 3.1 score is 9.2 (Critical) with AV:N/PR:N/UI:N but AC:H/AT:P, because only deployments that have not rotated the default secret and that still retain the demo tenant and admin account are affected. No public exploit identified at time of analysis, and no CISA KEV entry or vendor-patched version was provided in the source data.
RemediationAI
No vendor-released patch identified at time of analysis, so operators must mitigate directly. The primary control is to override the app_secret with a high-entropy, randomly generated value supplied via environment or deployment configuration instead of relying on the value in backend/app/config.py; this invalidates all previously forged tokens and any legitimately issued sessions signed with the old secret, forcing users to re-authenticate. Second, delete or disable the seeded 'admin' account and the 'tenant_demo' tenant in production deployments, since exploitation is scoped to that seeded administrator and tenant; removing them eliminates the specific token payload an attacker can mint, but verify no production data or credentials are stored under that tenant before deletion. Third, because the secret is public and static, restrict network exposure of the StaffDeck backend to trusted networks or place it behind an authenticating reverse proxy or VPN until a fix ships; this reduces reachability (the AV:N assumption) at the cost of breaking external or distributed access. Additionally, audit the tenant_demo stored model and channel credentials for exposure and rotate them, and monitor logs for API calls bearing tokens issued before the secret was rotated. Track the vendor advisory at https://www.vulncheck.com/advisories/openbmb-staffdeck-through-0.5.6-authentication-bypass-via-hard-coded-session-secret and the upstream repository for a patched release.
Affected VersionsENISA
Affected Packages
Source: OSV (Open Source Vulnerabilities)
Affects your stack
Affects 1 package in euvd
Checking this CVE against your own stacks and servers is part of the Basic plan and up.
Technical ContextAI · show details
The root cause is CWE-321 (Use of Hard-coded Cryptographic Key). StaffDeck's backend defines an application-level signing secret (app_secret) in backend/app/config.py, and the authentication module (backend/app/security/auth.py, roughly lines 96-98) uses that secret to produce and verify HMAC-SHA256 session tokens. Because the shipped default value is static and publicly visible in the upstream repository at tag v0.5.6, anyone who knows the default can compute the same HMAC over an arbitrary token payload. The advisory indicates the token payload can be crafted for the seeded 'admin' user inside the 'tenant_demo' tenant, so the signature check succeeds and the request is treated as an authenticated tenant administrator. The CPE entry cpe:2.3:a:openbmb:staffdeck:*:*:*:*:*:*:*:* is unbounded on version, so all releases at or before 0.5.6 should be treated as vulnerable; the exploitation primitive is entirely cryptographic/signature-based rather than a memory-safety or injection flaw, which is why AC:H and AT:P appear in the assessed vector.
Know when Staffdeck 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:H/AT:P/PR:N/UI:N/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N
Network-reachable token forgery needs no credentials (PR:N) but depends on the unchanged default secret and seeded demo tenant, folded into AC:H; full tenant-admin access yields C:H/I:H/A:H with no scope change.
Primary rating from Vendor (VulnCheck).
CVSS VectorVendor: VulnCheck
DescriptionCVE.org
OpenBMB StaffDeck through 0.5.6 contains an authentication bypass vulnerability that allows unauthenticated attackers to forge administrator session tokens using the hard-coded default app_secret. Attackers can sign an HMAC-SHA256 token for the seeded 'admin' user in 'tenant_demo' to gain full tenant administration, including stored model and channel credentials.
Executive ViewAI
OpenBMB StaffDeck releases up to 0.5.6 present an unauthenticated administrative-takeover risk for organizations that have left default secret material and demo tenant scaffolding in place, potentially exposing AI model and messaging channel credentials. Leadership should prioritize inventorying and hardening these deployments because the condition is configuration-dependent but the consequence-unauthorized tenant administration and credential theft-is severe if present. No public exploit identified at time of analysis, and no vendor-released patch identified at time of analysis.
No vendor-released patch identified at time of analysis. Interim mitigations: rotate the hard-coded default app_secret in backend/app/config.py, remove or disable the seeded 'admin' user and 'tenant_demo' tenant, restrict network access to OpenBMB StaffDeck interfaces, and monitor/audit for forged administrator session tokens. Apply vendor patch when available.
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.
EUVD-2026-103058
GHSA-m82j-2j7c-gx6q
NVD
CVE.org
EPSS
Vulnrichment
Exploit-DB