CVE-2025-12695VulnerabilityPublished 4 Nov 2025

CVE-2025-12695: dspy · Improper Isolation or Compartmentalization

Affects dspySource CVE Database V5

Not known to be exploited · CVSS 5.9 Medium.

Data as of 26 Oct 2026, 06:26 UTC

CVSS 3.1 base score

5.9Medium

How severe the flaw is on the 0–10 CVSS scale.

EPSS: 0.3% chance of exploitation in 30 days – higher than 26% of CVEsExploit Prediction Scoring System: the likelihood attackers will use this flaw soon.

CVSS:3.1/AV:N/AC:H/PR:N/UI:N/S:U/C:H/I:N/A:N

Attack Vector: Network
Attack Complexity: High
Privileges Required: None
User Interaction: None
Scope: Unchanged
Confidentiality: High
Integrity: None
Availability: None

Bold values are the worst case: easiest to exploit or largest impact.

Attack Vector: Network: The vulnerable component is bound to the network stack and the set of possible attackers extends beyond the other options listed, up to and including the entire Internet. Such a vulnerability is often termed 'remotely exploitable' and can be thought of as an attack being exploitable at the protocol level one or more network hops away (e.g., across one or more routers).
Attack Complexity: High: A successful attack depends on conditions beyond the attacker's control. That is, a successful attack cannot be accomplished at will, but requires the attacker to invest in some measurable amount of effort in preparation or execution against the vulnerable component before a successful attack can be expected. For example, a successful attack may require an attacker to: gather knowledge about the environment in which the vulnerable target/component exists; prepare the target environment to improve exploit reliability; or inject themselves into the logical network path between the target and the resource requested by the victim in order to read and/or modify network communications (e.g., a man in the middle attack).
Privileges Required: None: The attacker is unauthorized prior to attack, and therefore does not require any access to settings or files to carry out an attack.
User Interaction: None: The vulnerable system can be exploited without any interaction from any user.
Scope: Unchanged: An exploited vulnerability can only affect resources managed by the same security authority. In this case, the vulnerable component and the impacted component are either the same, or both are managed by the same security authority.
Confidentiality: High: There is total loss of confidentiality, resulting in all resources within the impacted component being divulged to the attacker. Alternatively, access to only some restricted information is obtained, but the disclosed information presents a direct, serious impact.
Integrity: None: There is no loss of integrity within the impacted component.
Availability: None: There is no impact to availability within the impacted component.

Definitions: FIRST CVSS v3.1 specification.

Exploitation evidence

Disclosed: evidence found.
CVE published 4 Nov 2025
Detectable: no evidence on record.
No public nuclei template indexed
Weaponised: no evidence on record.
No public exploit code indexed in Metasploit or Exploit-DB
Exploited in the wildExploited: no evidence on record.
Not in CISA KEV and no source reports exploitation
Observed by OffSeq MirageSeen by Mirage: no evidence on record.
No attack traffic recorded by OffSeq Mirage sensors

A missing step is not evidence of safety. Sources: cve.org (opens in a new tab), the Metasploit, Exploit-DB and nuclei-templates index as of 2026-10-26.

Disclosed: The CVE record is public on cve.org. Its publication is t = 0.
Detectable: A public nuclei template can find exposed instances, for defenders and attackers alike.
Weaponised: Public exploit code exists in Metasploit or Exploit-DB, or CISA SSVC reports a public proof of concept.
Exploited in the wild: CISA KEV lists it, CISA SSVC rates exploitation active, or a source reports attacks.
Observed by OffSeq Mirage: OffSeq Mirage honeypot sensors recorded attack traffic for this CVE, or found exposed hosts without attack traffic.

Moments are spaced evenly, not to scale; a dashed line marks a gap of six months or more. Deltas count from t = 0. Sources that give only a date (CISA KEV, Exploit-DB, or a time of exactly 00:00Z) read “same day” rather than hours. Metasploit modules carry the flaw’s disclosure date, not their release date, and nuclei templates carry no date, so both appear above but not on the timeline.

What to do

Affected software

Product	Affected	Fixed in
PyPIdspy`pkg:pypi/dspy`Other threats	Affected =0	Fixed in–

Check your own hosts for these packages with threat-finder (opens in a new tab), our free open-source scanner.

AI-suggested

To mitigate CVE-2025-12695, organizations should first audit their use of DSPy, especially AI agents that utilize the PythonInterpreter class. They should implement strict sandbox configurations that enforce least privilege and isolate interpreter execution environments to prevent unauthorized file system access. Employ containerization or virtual machines to further compartmentalize AI agent processes. Restrict or disable the use of PythonInterpreter in scenarios where user input is consumed unless absolutely necessary and secure. Monitor file access logs and sandbox escape attempts for suspicious activity. Apply network segmentation to limit exposure of AI development environments. Engage with DSPy maintainers for patches or updates addressing this vulnerability. Additionally, conduct regular security reviews of AI development pipelines and educate developers on secure sandboxing practices. If possible, implement runtime application self-protection (RASP) techniques to detect and block exploitation attempts in real-time.

Description

The overly permissive sandbox configuration in DSPy allows attackers to steal sensitive files in cases when users build an AI agent which consumes user input and uses the “PythonInterpreter” class.

Impact

For European organizations, the impact of CVE-2025-12695 centers on the potential theft of sensitive files, which could include intellectual property, personal data, or confidential business information. Organizations leveraging DSPy for AI agent development, particularly those integrating user input with PythonInterpreter execution, face increased risk of data breaches. This could lead to regulatory penalties under GDPR if personal data is exposed, reputational damage, and financial losses. The medium severity rating indicates that while exploitation is complex, the lack of required privileges and user interaction lowers barriers for attackers. Sectors such as finance, healthcare, and research institutions in Europe that rely on AI tools are particularly vulnerable. The breach of confidentiality could also undermine trust in AI deployments and delay innovation adoption. Additionally, the absence of known exploits suggests a window for proactive mitigation before active exploitation occurs.

CWE-653Improper Isolation or Compartmentalization (opens in a new tab)
Class weaknessOWASP Top 10 A06:2025: Insecure Design (opens in a new tab)
The product does not properly compartmentalize or isolate functionality, processes, or resources that require different privilege levels, rights, or permissions.
Architecture and Design
Break up privileges between different modules, objects, or entities. Minimize the interfaces between modules and require strong access control between them.
OWASP Cheat Sheets
Secure Design Principles (opens in a new tab)
From MITRE CWE, MITRE CAPEC and OWASP (Top 10, OpenCRE).

References

research.jfrog.com/vulnerabilities/dspy-sandbox-escape-arbitrary-file-read-jfsa-2025-001495652/ (opens in a new tab)

Community reviews

Loading reviews…

Exploitation evidence

What to do

Affected software

Description

AI analysisMachine-generated, updated 4 Nov 2025, 13:36 UTC

WeaknessesCWE-653

Technical details

References

Community reviews

Related threats