---METADATA---
VENDOR: MervinPraison
PRODUCT: PraisonAI
AFFECTED_VERSIONS: 0 up to (excluding) 1.6.78
CONFIDENCE: high
MISSING: none
---END_METADATA---

Description Summary:
PraisonAI contains a remote code execution vulnerability in CodeAgent._execute_python() that allows attackers to execute arbitrary code via LLM prompt injection.

Executive Summary:
An unauthenticated remote code execution vulnerability in PraisonAI allows attackers to fully compromise the host system by injecting malicious instructions into LLM-generated Python code.

Vulnerability Details

CVE-ID: CVE-2026-61447

Affected Software: MervinPraison PraisonAI

Affected Versions: 0 up to (excluding) 1.6.78

Vulnerability: This is a code injection vulnerability (CWE-94) stemming from the lack of AST validation or sandboxing in the CodeAgent._execute_python() function. An unauthenticated attacker can manipulate the LLM's output to execute arbitrary Python code on the underlying host.

Business Impact

Successful exploitation allows an attacker to gain full control over the host environment, including the exfiltration of sensitive environment secrets and the execution of arbitrary commands. Given the CVSS score of 10.0, this represents a critical risk that could lead to complete system compromise and significant data breaches.

Remediation Plan

Immediate Action: Upgrade PraisonAI to version 1.6.78 or later to incorporate necessary code validation and security restrictions.

Proactive Monitoring: Audit logs for unusual LLM interaction patterns or unexpected child process execution initiated by the PraisonAI service.

Compensating Controls: Ensure the application is deployed within a hardened, restricted-privilege container or sandbox environment to limit the impact of potential RCE.

Exploitation Status

Public Exploit Available: No

Analyst Notes: As of July 11, 2026, there is no public information indicating active exploitation or a public proof-of-concept for this vulnerability. The flaw's inherent exploitability is high due to the lack of input validation on generated code.

Analyst Recommendation

This vulnerability presents a maximum-severity risk to any deployment of PraisonAI. Organizations must prioritize updating to version 1.6.78 immediately to eliminate the possibility of remote code execution via prompt injection.

---METADATA---
VENDOR: MervinPraison
PRODUCT: PraisonAI
AFFECTED_VERSIONS: 0 up to (excluding) 1.6.78
CONFIDENCE: high
MISSING: none
---END_METADATA---

Description Summary:
PraisonAI is vulnerable to Code Injection via plugin auto-discovery, which may allow an unauthenticated attacker to execute arbitrary code on the host system.

Executive Summary:
A Code Injection vulnerability in PraisonAI can allow an attacker to execute arbitrary code on the underlying system, posing a critical risk to overall infrastructure security.

Vulnerability Details

CVE-ID: CVE-2026-61446

Affected Software: MervinPraison PraisonAI

Affected Versions: 0 up to (excluding) 1.6.78

Vulnerability: This vulnerability is a Code Injection (CWE-94) flaw related to plugin auto-discovery mechanisms. The system fails to properly control the generation of code, allowing an attacker to achieve Remote Code Execution.

Business Impact

This vulnerability allows for full system compromise, including the potential for data exfiltration, service disruption, and lateral movement within the network. With a CVSS score of 8.4, the risk of total system takeover makes this a critical security concern.

Remediation Plan

Immediate Action: Update PraisonAI to version 1.6.78 or later immediately to patch the code injection flaw.

Proactive Monitoring: Monitor system logs for unauthorized process execution or unexpected changes to the application environment.

Compensating Controls: Ensure the application is running with the least privilege necessary and isolate the instance using containers or virtualization to limit the scope of potential code execution.

Exploitation Status

Public Exploit Available: Unknown

Analyst Notes: As of July 17, 2026, there is no public information indicating active exploitation or a public proof-of-concept for this vulnerability. The nature of the flaw, involving code generation, makes it highly dangerous if an exploit is developed.

Analyst Recommendation

Due to the potential for total system compromise, this vulnerability requires immediate attention. Organizations must prioritize applying the update to version 1.6.78 to neutralize the risk of remote code execution.

---METADATA---
VENDOR: MervinPraison
PRODUCT: PraisonAI
AFFECTED_VERSIONS: 0 up to (excluding) 4.6.78
CONFIDENCE: high
MISSING: none
---END_METADATA---

Description Summary:
PraisonAI is vulnerable to arbitrary file write and command execution via the AICoder component due to missing path and command validation in LLM tool calls.

Executive Summary:
A critical vulnerability in the PraisonAI AICoder component allows authenticated attackers to perform arbitrary file writes and execute system commands with root privileges.

Vulnerability Details

CVE-ID: CVE-2026-61445

Affected Software: MervinPraison PraisonAI

Affected Versions: 0 up to (excluding) 4.6.78

Vulnerability: The vulnerability is a path traversal and command injection flaw (CWE-22) occurring within the AICoder component. It requires low-level authentication (PR:L) to interact with the chat interface, where an attacker can inject prompts that bypass path validation to write or execute files on the host filesystem.

Business Impact

An attacker who successfully exploits this flaw can achieve full system control by injecting commands that execute with root privileges. With a CVSS score of 9.9, this vulnerability poses a severe threat to the integrity and availability of the entire server environment.

Remediation Plan

Immediate Action: Update PraisonAI to version 4.6.78 or later to apply the necessary path validation and command sanitization patches.

Proactive Monitoring: Monitor filesystem activity for unexpected file creation or modification in sensitive directories, and review shell execution logs for anomalous activity.

Compensating Controls: Implement strict file system permissions and ensure the service runs with the least privilege required, preventing the application from executing commands as root.

Exploitation Status

Public Exploit Available: No

Analyst Notes: As of July 11, 2026, there is no public information indicating active exploitation or a public proof-of-concept for this vulnerability. The vulnerability is highly exploitable once an attacker has established a low-privileged session.

Analyst Recommendation

The potential for root-level command execution necessitates immediate patching. All users of PraisonAI must update to the latest version to prevent unauthorized system access and potential data destruction.

---METADATA---
VENDOR: MervinPraison
PRODUCT: PraisonAI
AFFECTED_VERSIONS: 0 up to (excluding) 4.6.78
CONFIDENCE: high
MISSING: none
---END_METADATA---

Description Summary:
PraisonAI contains a code injection vulnerability in its API module where unsanitized user input is passed to subprocess.Popen(), allowing arbitrary code execution.

Executive Summary:
A critical code injection vulnerability in PraisonAI allows authenticated attackers to achieve remote code execution on the underlying host system.

Vulnerability Details

CVE-ID: CVE-2026-61444

Affected Software: MervinPraison PraisonAI

Affected Versions: 0 up to (excluding) 4.6.78

Vulnerability: The application is susceptible to CWE-94 (Code Injection) because the agents_file parameter in deploy/api.py is interpolated into an f-string without sanitization. This allows an authenticated user to inject malicious Python code that is subsequently executed by the server via subprocess.Popen().

Business Impact

The successful exploitation of this vulnerability grants an attacker full control over the application server, leading to potential data exfiltration, system compromise, and lateral movement within the network. Given the CVSS score of 9.1, this flaw represents a significant risk to the integrity and availability of the infrastructure.

Remediation Plan

Immediate Action: Upgrade PraisonAI to version 4.6.78 or later immediately to incorporate the necessary input sanitization patches.

Proactive Monitoring: Review system and application logs for suspicious subprocess calls or unexpected file modifications originating from the API service.

Compensating Controls: Implement strict network access control lists (ACLs) to restrict access to the API deployment endpoint to only known, trusted administrative IP addresses.

Exploitation Status

Public Exploit Available: False

Analyst Notes: As of Jul 10, 2026, there is no public information indicating active exploitation of this vulnerability. However, due to the nature of the flaw, the potential for exploitation is high.

Analyst Recommendation

This vulnerability presents a severe risk due to the potential for full system compromise. Administrators must prioritize updating the PraisonAI software to the patched version as the primary defense against potential code injection attacks.

---METADATA---
VENDOR: MervinPraison
PRODUCT: PraisonAI
AFFECTED_VERSIONS: 0 up to (excluding) 0.1.9
CONFIDENCE: high
MISSING: none
---END_METADATA---

Description Summary:
PraisonAI Platform is vulnerable to a missing authorization flaw allowing logged-in users to perform unauthorized actions due to insufficient access control checks.

Executive Summary:
A missing authorization vulnerability in the PraisonAI platform allows authenticated attackers to manipulate system data, posing a significant risk to data integrity.

Vulnerability Details

CVE-ID: CVE-2026-61442

Affected Software: MervinPraison PraisonAI

Affected Versions: 0 up to (excluding) 0.1.9

Vulnerability: The application suffers from a missing authorization vulnerability (CWE-862). The CVSS vector (PR:L) confirms that an attacker must possess low-level user privileges to exploit this flaw, which enables them to perform unauthorized operations that should be restricted.

Business Impact

The vulnerability carries a CVSS score of 7.1, indicating a high severity. Successful exploitation could lead to unauthorized modification of sensitive data or unauthorized system operations, potentially resulting in data corruption or a loss of trust in the platform's security controls.

Remediation Plan

Immediate Action: Upgrade to PraisonAI version 0.1.9 or later to incorporate the necessary authorization checks.

Proactive Monitoring: Review application logs for unusual API calls or actions performed by low-privileged accounts that deviate from standard user behavior.

Compensating Controls: Implement strict identity and access management (IAM) policies to limit the potential impact of compromised user accounts while the update is being deployed.

Exploitation Status

Public Exploit Available: Unknown

Analyst Notes: As of July 13, 2026, there is no public information indicating active exploitation or a public proof-of-concept for this vulnerability. The flaw is primarily an access control implementation error, which is highly exploitable once an attacker has gained a foothold as a standard user.

Analyst Recommendation

Given the high CVSS score and the nature of authorization bypasses, organizations should prioritize upgrading the PraisonAI platform to version 0.1.9 immediately. Applying this patch is the only definitive way to enforce proper authorization constraints and mitigate the risk of unauthorized data manipulation.

---METADATA---
VENDOR: MervinPraison
PRODUCT: PraisonAI
AFFECTED_VERSIONS: 0 up to (excluding) 4.6.78
CONFIDENCE: high
MISSING: none
---END_METADATA---

Description Summary:
PraisonAI versions before 4.6.78 are susceptible to a vulnerability related to insecure default resource initialization, facilitating potential security bypasses.

Executive Summary:
An insecure default initialization vulnerability in PraisonAI versions prior to 4.6.78 poses a high risk by allowing potential security bypasses.

Vulnerability Details

CVE-ID: CVE-2026-61439

Affected Software: MervinPraison PraisonAI

Affected Versions: 0 up to (excluding) 4.6.78

Vulnerability: The software suffers from CWE-1188, where resources are initialized with insecure defaults, allowing unauthenticated attackers to bypass prompt injection defenses.

Business Impact

The ability to bypass security defenses in an AI-driven platform can lead to the manipulation of model outputs or unauthorized access to integrated systems. With a CVSS score of 7.5, this vulnerability represents a high risk to the integrity of the AI's decision-making process and overall system security.

Remediation Plan

Immediate Action: Update PraisonAI to version 4.6.78 or later.

Proactive Monitoring: Monitor AI interaction logs for patterns consistent with injection attempts or unexpected system behavior.

Compensating Controls: Apply strict input validation and sanitization layers in front of the PraisonAI interface to mitigate potential prompt injection attacks.

Exploitation Status

Public Exploit Available: No

Analyst Notes: As of July 12, 2026, there is no public information indicating active exploitation or a public proof-of-concept for this vulnerability. The vulnerability requires immediate attention to prevent malicious manipulation of AI agents.

Analyst Recommendation

Given the potential for security control bypass, it is critical that users update to the latest patched version. Organizations should treat this as a high-priority update to maintain the security posture of their AI deployments.

---METADATA---
VENDOR: MervinPraison
PRODUCT: PraisonAI
AFFECTED_VERSIONS: 0 up to (excluding) 1.6.78
CONFIDENCE: high
MISSING: none
---END_METADATA---

Description Summary:
PraisonAI is vulnerable to a protection mechanism failure that can lead to remote code execution.

Executive Summary:
A critical protection mechanism failure in MervinPraison PraisonAI allows for potential remote code execution, posing a significant risk to host system integrity.

Vulnerability Details

CVE-ID: CVE-2026-61437

Affected Software: MervinPraison PraisonAI

Affected Versions: 0 up to (excluding) 1.6.78

Vulnerability: The vulnerability is identified as a protection mechanism failure (CWE-693). It allows an attacker to bypass security controls, resulting in high-impact compromise of system confidentiality, integrity, and availability.

Business Impact

With a CVSS score of 7.8, this vulnerability carries a High severity rating. Successful exploitation could grant an attacker full control over the host environment where the PraisonAI package is deployed, potentially leading to unauthorized data access, lateral movement, or complete system takeover.

Remediation Plan

Immediate Action: Update the praisonaiagents pip package to version 1.6.78 or later immediately.

Proactive Monitoring: Review system and application logs for unusual process execution or unauthorized file access attempts originating from the PraisonAI service account.

Compensating Controls: Ensure the application is running with the principle of least privilege, ideally within an isolated container or virtual environment to restrict potential impact.

Exploitation Status

Public Exploit Available: false

Analyst Notes: As of July 11, 2026, there is no public information indicating active exploitation of this vulnerability. However, due to the nature of the flaw, the potential for exploitation is high.

Analyst Recommendation

Given the potential for remote code execution, this vulnerability represents a significant threat to your infrastructure. Administrators should prioritize the update to version 1.6.78 across all deployments immediately to mitigate the risk of compromise.

---METADATA---
VENDOR: MervinPraison
PRODUCT: PraisonAI
AFFECTED_VERSIONS: 0 up to (excluding) 4.6.78
CONFIDENCE: high
MISSING: none
---END_METADATA---

Description Summary:
PraisonAI is vulnerable to an authentication bypass due to missing webhook signature verification, allowing unauthenticated attackers to trigger actions.

Executive Summary:
A critical authentication bypass vulnerability in PraisonAI, caused by missing webhook signature verification, allows unauthenticated attackers to perform unauthorized operations.

Vulnerability Details

CVE-ID: CVE-2026-61436

Affected Software: MervinPraison PraisonAI

Affected Versions: 0 up to (excluding) 4.6.78

Vulnerability: The software fails to implement proper webhook signature verification, which constitutes an improper authentication vulnerability. This allows an unauthenticated attacker to send forged requests that the system accepts as legitimate.

Business Impact

An attacker can exploit this flaw to execute unauthorized commands or trigger workflows within the PraisonAI environment. Given the CVSS score of 8.6, this poses a significant risk to the integrity and availability of the application, as unauthorized state changes can be forced without any administrative credentials.

Remediation Plan

Immediate Action: Update PraisonAI to version 4.6.78 or later to ensure proper webhook signature verification is enforced.

Proactive Monitoring: Inspect webhook request logs for incoming traffic that lacks valid signatures or originates from unauthorized IP ranges.

Compensating Controls: If an immediate update is not feasible, restrict access to the webhook endpoint at the network level to known, trusted sources only.

Exploitation Status

Public Exploit Available: Unknown

Analyst Notes: As of July 17, 2026, there is no confirmed active exploitation in the wild; however, per CISA's SSVC assessment a proof-of-concept exists, so exploitation risk should be treated as credible. The lack of signature validation makes this an easily exploitable target for attackers.

Analyst Recommendation

The severity of this authentication bypass requires immediate patching to prevent unauthorized command execution. Administrators should move to version 4.6.78 immediately to restore secure webhook processing.

---METADATA---
VENDOR: MervinPraison
PRODUCT: PraisonAI
AFFECTED_VERSIONS: 0 up to (excluding) 4.6.78
CONFIDENCE: high
MISSING: none
---END_METADATA---

Description Summary:
PraisonAI contains an authentication bypass vulnerability via host header spoofing, allowing unauthenticated attackers to gain unauthorized access.

Executive Summary:
An authentication bypass vulnerability in PraisonAI versions prior to 4.6.78 allows unauthenticated attackers to compromise system integrity via host header spoofing.

Vulnerability Details

CVE-ID: CVE-2026-61435

Affected Software: MervinPraison PraisonAI

Affected Versions: 0 up to (excluding) 4.6.78

Vulnerability: This vulnerability involves improper authentication, allowing unauthenticated attackers to bypass security checks through host header spoofing.

Business Impact

Successful exploitation allows an attacker to bypass authentication, potentially leading to full control over the application and its underlying data. With a CVSS score of 8.2, this flaw poses a significant risk to the overall security posture and data integrity of the PraisonAI platform.

Remediation Plan

Immediate Action: Update PraisonAI to version 4.6.78 or later to patch the authentication bypass vulnerability.

Proactive Monitoring: Audit application logs for suspicious host headers or unauthorized requests that bypass typical login workflows.

Compensating Controls: Deploy a Web Application Firewall (WAF) configured to validate and filter host headers to prevent spoofing attempts.

Exploitation Status

Public Exploit Available: No confirmed public exploit in the available data.

Analyst Notes: As of July 17, 2026, there is no confirmed active exploitation in the wild; however, per CISA's SSVC assessment a proof-of-concept exists, so exploitation risk should be treated as credible. The vulnerability is inherently critical due to the ease of bypassing authentication via header manipulation.

Analyst Recommendation

Given the availability of a proof-of-concept and the high severity of an authentication bypass, immediate patching is required. Administrators should upgrade to version 4.6.78 immediately to neutralize this threat.

---METADATA---
VENDOR: MervinPraison
PRODUCT: PraisonAI
AFFECTED_VERSIONS: 0 up to (excluding) 4.6.78
CONFIDENCE: high
MISSING: none
---END_METADATA---

Description Summary:
PraisonAI contains an OS command injection vulnerability allowing authenticated attackers to execute arbitrary system commands via a bypass.

Executive Summary:
A critical OS command injection vulnerability in MervinPraison PraisonAI allows authenticated attackers to execute arbitrary system commands.

Vulnerability Details

CVE-ID: CVE-2026-61434

Affected Software: MervinPraison PraisonAI

Affected Versions: 0 up to (excluding) 4.6.78

Vulnerability: This vulnerability (CWE-78) exists due to an allowlist bypass, which allows an authenticated attacker to execute OS commands via find-exec. By bypassing existing input filters, an attacker can gain unauthorized execution capabilities on the host.

Business Impact

The CVSS score of 8.8 highlights the critical nature of this vulnerability, which grants authenticated users the ability to execute commands on the underlying server. This could lead to a complete system takeover, data exfiltration, or the deployment of persistent malicious agents within the environment.

Remediation Plan

Immediate Action: Update PraisonAI to version 4.6.78 or later to resolve the allowlist bypass and prevent command injection.

Proactive Monitoring: Audit application logs for suspicious command execution or unexpected usage of system utilities like find or exec.

Compensating Controls: Ensure the application runs with the lowest possible system privileges and utilize a WAF to inspect incoming requests for suspicious shell metacharacters.

Exploitation Status

Public Exploit Available: false

Analyst Notes: As of July 11, 2026, there is no public information indicating active exploitation of this vulnerability. However, due to the nature of the flaw, the potential for exploitation is high.

Analyst Recommendation

This vulnerability represents a severe security risk that could lead to full host compromise. Administrators must prioritize updating PraisonAI to version 4.6.78 to remediate the OS command injection flaw and secure the application against unauthorized command execution.

---METADATA---
VENDOR: MervinPraison
PRODUCT: PraisonAI
AFFECTED_VERSIONS: 0 up to (excluding) 1.6.78
CONFIDENCE: high
MISSING: none
---END_METADATA---

Description Summary:
PraisonAI is vulnerable to Server-Side Request Forgery (SSRF) due to improper input validation, allowing authenticated attackers to interact with internal resources.

Executive Summary:
A Server-Side Request Forgery vulnerability in PraisonAI allows authenticated attackers to perform unauthorized requests, posing a significant risk to internal network security.

Vulnerability Details

CVE-ID: CVE-2026-61430

Affected Software: MervinPraison PraisonAI

Affected Versions: 0 up to (excluding) 1.6.78

Vulnerability: This flaw involves a Server-Side Request Forgery (CWE-918) vulnerability. An attacker with low-level authenticated access can manipulate the application to send requests to unintended internal destinations.

Business Impact

The ability to perform SSRF can lead to unauthorized access to internal services, sensitive data exfiltration, or the discovery of internal network infrastructure. Given the CVSS score of 8.5, this high-severity vulnerability represents a significant risk to the integrity and confidentiality of the internal environment.

Remediation Plan

Immediate Action: Upgrade to PraisonAI version 1.6.78 or later to resolve the underlying SSRF flaw.

Proactive Monitoring: Monitor network traffic originating from the application server for unexpected outbound connections to internal IP addresses or sensitive endpoints.

Compensating Controls: Implement strict egress filtering on the application server to restrict outbound traffic to only necessary and trusted external services.

Exploitation Status

Public Exploit Available: Unknown

Analyst Notes: As of July 17, 2026, there is no confirmed active exploitation in the wild; however, per CISA's SSVC assessment a proof-of-concept exists, so exploitation risk should be treated as credible. The vulnerability is inherently dangerous as it allows attackers to bypass network perimeters.

Analyst Recommendation

This vulnerability presents a high risk due to the potential for unauthorized internal network access. Organizations using PraisonAI should prioritize updating to the patched version immediately to prevent potential exploitation of this SSRF flaw.

---METADATA---
VENDOR: MervinPraison
PRODUCT: PraisonAI
AFFECTED_VERSIONS: 0 up to (excluding) 1.6.78
CONFIDENCE: high
MISSING: none
---END_METADATA---

Description Summary:
A Server-Side Request Forgery (SSRF) vulnerability in PraisonAI versions before 1.6.78 allows authenticated users to trigger unauthorized requests to internal network resources.

Executive Summary:
An authenticated SSRF vulnerability in PraisonAI allows attackers to leverage the application to perform unauthorized requests against internal network services.

Vulnerability Details

CVE-ID: CVE-2026-61429

Affected Software: MervinPraison PraisonAI

Affected Versions: 0 up to (excluding) 1.6.78

Vulnerability: This vulnerability (CWE-918) occurs within the crawl4ai chromium backend integration. An authenticated attacker can manipulate the application to send requests to arbitrary internal services, potentially bypassing network perimeter security.

Business Impact

With a CVSS score of 8.5, this SSRF vulnerability enables an attacker to interact with internal services that are otherwise unreachable from the internet. This can lead to the discovery of internal network topology, exploitation of non-public services, or the extraction of internal metadata, posing a critical risk to internal infrastructure security.

Remediation Plan

Immediate Action: Update PraisonAI to version 1.6.78 or later to patch the SSRF vulnerability in the chromium backend.

Proactive Monitoring: Monitor network egress traffic from the application server for unusual requests targeting internal IP ranges or restricted services.

Compensating Controls: Use network-level egress filtering to restrict the application server's ability to communicate with sensitive internal assets.

Exploitation Status

Public Exploit Available: Unknown.

Analyst Notes: As of July 12, 2026, there is no public information indicating active exploitation or a published weaponized exploit for this vulnerability. The vulnerability is highly dependent on the attacker having authenticated access to the application.

Analyst Recommendation

Users should update to version 1.6.78 immediately to address this SSRF risk. Given the potential for attackers to pivot into internal networks through this flaw, maintaining strict egress controls alongside the software update is strongly recommended to protect internal resources.

---METADATA---
VENDOR: MervinPraison
PRODUCT: PraisonAI
AFFECTED_VERSIONS: MervinPraison PraisonAI: 0 up to (excluding) 4.6.78
CONFIDENCE: high
MISSING: none
---END_METADATA---

Description Summary:
An authentication bypass vulnerability in PraisonAI AgentMail allows remote attackers to perform unauthorized message injection via webhooks.

Executive Summary:
PraisonAI AgentMail contains an authentication bypass flaw that permits unauthenticated attackers to inject messages via webhooks, risking data integrity.

Vulnerability Details

CVE-ID: CVE-2026-61428

Affected Software: MervinPraison PraisonAI

Affected Versions: 0 up to 4.6.78

Vulnerability: The software is susceptible to CWE-290 (Authentication Bypass by Spoofing), where an unauthenticated attacker can manipulate webhook inputs to inject malicious messages into the system.

Business Impact

This vulnerability allows for the unauthorized injection of messages, which could be used to manipulate AI-driven workflows or deceive users. With a CVSS score of 7.3, the potential for data tampering and loss of trust in automated systems is high, necessitating prompt remediation to prevent integrity-based attacks.

Remediation Plan

Immediate Action: Update PraisonAI to version 4.6.78 or later to incorporate the vendor-supplied fix.

Proactive Monitoring: Monitor system logs for suspicious webhook activity or unexpected message patterns that deviate from standard operational behavior.

Compensating Controls: Utilize a Web Application Firewall (WAF) to inspect and filter incoming webhook requests, specifically looking for anomalous headers or payload structures.

Exploitation Status

Public Exploit Available: Unknown.

Analyst Notes: As of July 12, 2026, there is no public information indicating active exploitation or a published proof-of-concept for this vulnerability. The flaw is fundamentally an input validation issue that is effectively mitigated by the provided patch.

Analyst Recommendation

Organizations utilizing PraisonAI should treat this vulnerability as a high priority for remediation. Updating to the latest version is the only robust method to eliminate the authentication bypass vector and ensure the integrity of your AI messaging environment.

---METADATA---
VENDOR: MervinPraison
PRODUCT: PraisonAI
AFFECTED_VERSIONS: 0 up to (excluding) 4.6.78
CONFIDENCE: high
MISSING: none
---END_METADATA---

Description Summary:
PraisonAI contains an unauthenticated Server-Side Request Forgery (SSRF) vulnerability that allows attackers to force the server to send requests to arbitrary internal or external destinations.

Executive Summary:
An unauthenticated Server-Side Request Forgery (SSRF) flaw in PraisonAI allows attackers to trigger unauthorized outbound requests from the application server.

Vulnerability Details

CVE-ID: CVE-2026-60091

Affected Software: MervinPraison PraisonAI

Affected Versions: 0 up to (excluding) 4.6.78

Vulnerability: This vulnerability is a Server-Side Request Forgery (CWE-918) that allows unauthenticated attackers to supply malicious inputs, forcing the application to perform unauthorized requests. This can be used to scan internal networks or interact with internal services that are otherwise unreachable from the public internet.

Business Impact

With a CVSS score of 7.2 (High), this SSRF vulnerability represents a significant risk to internal infrastructure. Attackers can bypass firewall protections to reach internal APIs, cloud metadata services, or other sensitive infrastructure, leading to potential data exfiltration or service disruption.

Remediation Plan

Immediate Action: Update PraisonAI to version 4.6.78 or later to patch the SSRF vulnerability.

Proactive Monitoring: Monitor outbound network traffic from the PraisonAI server for connections to sensitive internal IP ranges or unexpected external domains.

Compensating Controls: Implement egress filtering at the network level to restrict the server's ability to communicate with internal resources that it does not explicitly need to access.

Exploitation Status

Public Exploit Available: Unknown

Analyst Notes: As of July 12, 2026, there is no confirmed active exploitation in the wild; however, per CISA's SSVC assessment a proof-of-concept exists, so exploitation risk should be treated as credible. The unauthenticated nature of this flaw makes it highly accessible to remote attackers.

Analyst Recommendation

The combination of unauthenticated access and SSRF-based internal network exposure makes this vulnerability a high priority for remediation. Administrators are urged to update to version 4.6.78 immediately to prevent potential exploitation of internal systems.

---METADATA---
VENDOR: MervinPraison
PRODUCT: PraisonAI
AFFECTED_VERSIONS: 0 up to (excluding) 4.6.78
CONFIDENCE: high
MISSING: none
---END_METADATA---

Description Summary:
PraisonAI fails to validate the dimension argument in its knowledge-store backends, enabling SQL/CQL injection via specially crafted strings.

Executive Summary:
An unauthenticated SQL/CQL injection vulnerability in PraisonAI allows attackers to execute arbitrary database commands via the knowledge-store API.

Vulnerability Details

CVE-ID: CVE-2026-60090

Affected Software: MervinPraison PraisonAI

Affected Versions: 0 to 4.6.77

Vulnerability: This is an SQL injection vulnerability (CWE-89) where the 'dimension' argument is improperly neutralized. Attackers can inject malicious SQL/CQL tokens into the database statement, leading to unauthorized data manipulation or deletion.

Business Impact

The vulnerability allows for total control over database operations, including the ability to drop tables or exfiltrate sensitive information. With a CVSS score of 9.8, the potential for data loss and severe operational disruption is critical, necessitating immediate remediation.

Remediation Plan

Immediate Action: Update PraisonAI to version 4.6.78 or later to incorporate input validation fixes.

Proactive Monitoring: Monitor database query logs for anomalous syntax, specifically looking for injected SQL tokens like DROP, UNION, or comment syntax (--).

Compensating Controls: Utilize database-level permissions to restrict the application user's access, ensuring the service account cannot drop tables or access sensitive schemas.

Exploitation Status

Public Exploit Available: Unknown.

Analyst Notes: As of Jul 11, 2026, there is no public information indicating active exploitation or a public proof-of-concept for this vulnerability. The flaw is highly exploitable due to the lack of input sanitization in the database query generation.

Analyst Recommendation

Given the ease of exploitation and the severity of SQL injection, upgrading to the patched version is essential. Verify that the update has been applied successfully and audit database logs for any signs of prior unauthorized access attempts.

---METADATA---
VENDOR: MervinPraison
PRODUCT: praisonai-platform
AFFECTED_VERSIONS: < 0.1.4
CONFIDENCE: high
MISSING: none
---END_METADATA---

Description Summary:
The PraisonAI platform is susceptible to authorization bypass and missing authorization checks, allowing authenticated users to perform unauthorized actions.

Executive Summary:
An authorization bypass vulnerability in the MervinPraison praisonai-platform allows authenticated users to exceed their intended privilege levels and access restricted system functions.

Vulnerability Details

CVE-ID: CVE-2026-48169

Affected Software: MervinPraison praisonai-platform

Affected Versions: < 0.1.4

Vulnerability: The application suffers from missing authorization (CWE-862) and authorization bypass via user-controlled keys (CWE-639), which can be exploited by an authenticated attacker to manipulate system resources.

Business Impact

With a CVSS score of 8.8, this vulnerability presents a significant risk to data integrity and system control. Authenticated users could potentially escalate their privileges or access sensitive multi-agent configurations, leading to unauthorized data modification or total system compromise.

Remediation Plan

Immediate Action: Update the praisonai-platform to version 0.1.4 or higher to resolve the authorization flaws.

Proactive Monitoring: Monitor application logs for unusual API calls or actions performed by non-privileged accounts that deviate from standard user behavior.

Compensating Controls: Implement strict role-based access control (RBAC) at the network or application gateway level to limit the exposure of sensitive endpoints to authenticated users only.

Exploitation Status

Public Exploit Available: Unknown.

Analyst Notes: As of August 8, 2026, there is no public information indicating active exploitation or a public proof-of-concept for this vulnerability. The requirement for a pre-existing authenticated session slightly reduces the attack surface compared to unauthenticated vulnerabilities.

Analyst Recommendation

Security teams must prioritize updating the PraisonAI platform to the specified fixed version. Ensuring that authorization checks are properly enforced is critical to maintaining the integrity of multi-agent systems and preventing unauthorized administrative control.