Cyber Resilience

CVE-2024-9099

Lunary 1.4.29

Public PoC
Published
20 March 2025
Modified
10 April 2025
Patch / advisory
View advisory
CVSS Score v3.1 8.1
Click a component to see what it means
Raw vectorCVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:H/A:N
EPSS Score 0.0055 43th percentile
Risk Priority 60 floored blend · peak EPSS

Summary

CVE-2024-9099 is a high-severity Exposure of Sensitive Information Through Metadata (CWE-1230) vulnerability in Lunary Lunary. Its CVSS base score is 8.1 (High).

Operationally, exploitation aligns with the MITRE ATT&CK technique Data from Local System (T1005); ranked at the 43th percentile by exploit likelihood (below the median); it is not currently listed in the CISA KEV catalog; a public proof-of-concept is referenced.

This vulnerability is AI-related — categorised as LLM Application Platforms; in the Privacy and Disclosure risk domain.

The strongest mitigations our analysis identified map to AC-4 (Information Flow Enforcement) and AC-16 (Security and Privacy Attributes) — see the control section below for these in your framework.

Deeper analysis AI-assisted summary

Synthesised by an AI model from the NVD description and linked references — a reading aid, not an authoritative source.

CVE-2024-9099 is a vulnerability in lunary-ai/lunary version v1.4.29, where the GET /projects API endpoint exposes both public and private API keys for all projects to users with minimal permissions, such as Viewers or Prompt Editors. The private API keys become visible in the developer tools when the endpoint is called from the frontend, enabling unauthorized access to sensitive credentials.

Attackers require only low-privilege access (PR:L), such as a Viewer or Prompt Editor role, to exploit this remotely (AV:N) with low attack complexity (AC:L) and no user interaction (UI:N), earning a CVSS v3.1 base score of 8.1 (C:H/I:H/A:N). Successful exploitation allows retrieval of the API keys, which can then be used to perform actions on behalf of the affected project, access private data, and delete resources.

A fix is available via commit 8ba1b8ba2c2c30b1cec30eb5777c1fda670cbbfc in the lunary-ai/lunary GitHub repository. Further details on the vulnerability and remediation are provided on the Huntr bounty page at https://huntr.com/bounties/ffb84fe8-3e60-4200-ac2d-1fd1e1c93e91.

EU & UK References

Vulnerability Data

In lunary-ai/lunary version v1.4.29, the GET /projects API endpoint exposes both public and private API keys for all projects to users with minimal permissions, such as Viewers or Prompt Editors. This vulnerability allows unauthorized users to retrieve sensitive credentials, which…

more

can be used to perform actions on behalf of the project, access private data, and delete resources. The private API keys are exposed in the developer tools when the endpoint is called from the frontend.

CWE(s)
CWE-1230NVD-CWE-Other

AI Security AnalysisAI

AI Category
LLM Application Platforms
Risk Domain
Privacy and Disclosure
OWASP Top 10 for LLMs 2025
None mapped
Classification Reason
Matched keywords: ai, lunary

Related Threats

MITRE ATT&CK Enterprise Techniques

T1005 Data from Local System Collection
Adversaries may search local system sources, such as file systems, configuration files, local databases, virtual machine files, or process memory, to find files of interest and sensitive data prior to Exfiltration.
attack.mitre.org →
T1025 Data from Removable Media Collection
Adversaries may search connected removable media on computers they have compromised to find files of interest.
attack.mitre.org →
T1083 File and Directory Discovery Discovery
Adversaries may enumerate files and directories or may search in specific locations of a host or network share for certain information within a file system.
attack.mitre.org →
T1217 Browser Information Discovery Discovery
Adversaries may enumerate information about browsers to learn more about compromised environments.
attack.mitre.org →
T1592 Gather Victim Host Information Reconnaissance
Adversaries may gather information about the victim's hosts that can be used during targeting.
attack.mitre.org →
Derived from this CVE’s CWE(s) via the direct CWE→ATT&CK cross-walk.

CVEs Like This One

CVE-2024-8765Same product: Lunary Lunary
CVE-2024-5133Same product: Lunary Lunary
CVE-2024-1738Same product: Lunary Lunary
CVE-2024-5386Same product: Lunary Lunary
CVE-2024-6086Same product: Lunary Lunary
CVE-2024-5130Same product: Lunary Lunary
CVE-2024-10272Same product: Lunary Lunary
CVE-2024-1902Same product: Lunary Lunary
CVE-2024-5277Same product: Lunary Lunary
CVE-2024-8999Same product: Lunary Lunary

Affected Assets

lunary
lunary
1.4.29

Mitigating Controls

Control response

Prevent
Stop it (NIST 800-53)

—

Detect
Catch it (NIST detect / respond)

—

Harden
Shrink the surface (DISA STIG)
  • 1 hardening rule · 1 OS baseline
Validate
Prove the fix (OWASP ASVS)
  • V3.4.2
  • V12.1.5
  • V13.3.2
  • V14.2.4

Mitigating Controls (NIST 800-53 r5) AI

AC-4 Information Flow Enforcement good match

Information flow enforcement can be configured to treat metadata as protected information and block its unauthorized release.

AC-16 Security and Privacy Attributes partial match

Associating and enforcing security/privacy attributes on objects directly governs what metadata may be exposed.

SC-16 Transmission of Security and Privacy Attributes partial match

Requiring security attributes to accompany transmitted data prevents sensitive values from leaking through metadata channels.

Mitigating Controls (NIST CSF 2.0) AI

Derived directly from the weakness types (CWEs) cited in the NVD entry via our AI-authored CWE→CSF cross-walk (authority under review) — links open the control.

ID.AM-07 mostly match
prevents

Explicit inventorying of metadata enables identification and subsequent protection of sensitive derived information.

PR.AA-05 mostly match
prevents

Least-privilege access policies applied to metadata directly prevent unauthorized exposure of sensitive derived data.

Mitigating Controls (ISO/IEC 27001:2022 Annex A) AI

Derived directly from the weakness types (CWEs) cited in the NVD entry via our AI-authored CWE→ISO cross-walk (authority under review) — links open the control.

A.5.12 Classification of information partial match
mitigates

Classification helps identify metadata that may need protection, but does not directly limit its exposure.

A.5.13 Labelling of information partial match
mitigates

Labelling can flag sensitive metadata, yet does not enforce technical controls to prevent its disclosure.

A.5.14 Information transfer partial match
degrades

Transfer policies can require stripping or protecting metadata, but coverage is indirect.

A.5.15 Access control partial match
mitigates

Access-control rules can be extended to metadata objects, though the control is not metadata-specific.

A.8.11 Data masking partial match
mitigates

Data-masking techniques can obscure sensitive metadata values, providing partial mitigation.

A.8.12 Data leakage prevention partial match
finds

DLP solutions can detect and block metadata leakage, yet are not designed solely for this weakness.

References