PT-2026-94052

Name of the Vulnerable Software and Affected Versions Wiki.js versions prior to 2.5.315 Description Authorization checks in multiple GraphQL resolvers omit page tags, which allows the bypass of tag-based access restrictions. An attacker can use the list, tree, tags, searchTags, and links resolver...

GHSA-5648-RGJ9-V224 @zereight/mcp-gitlab has multiple safety-control bypasses: execute_graphql read-only + allow-list bypass, unauthenticated transports, session-exhaustion DoS

Summary @zereight/mcp-gitlab exposes GitLab to an LLM agent while relying on read-only mode, a project allow-list, and transport auth as its safety controls. Five defects defeat those controls. Under the MCP threat model, tool-call arguments/content can be shaped by untrusted input prompt injecti...

@zereight/mcp-gitlab has multiple safety-control bypasses: execute_graphql read-only + allow-list bypass, unauthenticated transports, session-exhaustion DoS

Summary @zereight/mcp-gitlab exposes GitLab to an LLM agent while relying on read-only mode, a project allow-list, and transport auth as its safety controls. Five defects defeat those controls. Under the MCP threat model, tool-call arguments/content can be shaped by untrusted input prompt injecti...

CVE-2026-76820

OpenCTI is an open source platform for managing cyber threat intelligence knowledge and observables. Prior to 7.260701.0, the synchronizerFetch GraphQL query called fetchRemoteStreams after checking only that a remote stream URL used HTTP or HTTPS. The backend did not apply the ingestion deny lis...

CVE-2026-76821

OpenCTI is an open source platform for managing cyber threat intelligence knowledge and observables. Prior to 7.260706.0, the JSON ingestion mapper's extractWithRegexp formula function compiled a user-supplied regular expression with the JavaScript RegExp engine in...

CVE-2026-76820 OpenCTI: Synchronizer SSRF: stream fetch has no URL validation

OpenCTI is an open source platform for managing cyber threat intelligence knowledge and observables. Prior to 7.260701.0, the synchronizerFetch GraphQL query called fetchRemoteStreams after checking only that a remote stream URL used HTTP or HTTPS. The backend did not apply the ingestion deny lis...

EUVD-2026-79241

OpenCTI is an open source platform for managing cyber threat intelligence knowledge and observables. Prior to 7.260701.0, the synchronizerFetch GraphQL query called fetchRemoteStreams after checking only that a remote stream URL used HTTP or HTTPS. The backend did not apply the ingestion deny lis...

CVE-2026-76820

OpenCTI, an open-source cyber threat intelligence platform, is affected by a Server-Side Request Forgery (SSRF) in the synchronizerFetch GraphQL query prior to version 7.260701.0 . The backend's fetchRemoteStreams function validated only the HTTP/HTTPS scheme without applying the ingestion deny l...

EUVD-2026-79240

OpenCTI is an open source platform for managing cyber threat intelligence knowledge and observables. Prior to 7.260706.0, the JSON ingestion mapper's extractWithRegexp formula function compiled a user-supplied regular expression with the JavaScript RegExp engine in...

CVE-2026-76821 OpenCTI: User-Controlled ReDoS in JSON Ingestion Mapper

OpenCTI is an open source platform for managing cyber threat intelligence knowledge and observables. Prior to 7.260706.0, the JSON ingestion mapper's extractWithRegexp formula function compiled a user-supplied regular expression with the JavaScript RegExp engine in...

CVE-2026-76821

OpenCTI , an open-source cyber threat intelligence platform, is vulnerable to user-controlled ReDoS in the JSON ingestion mapper's extractWithRegexp formula function (prior to 7.260706.0 ). The function compiled a user-supplied regular expression with the JavaScript RegExp engine without validati...

CVE-2026-76821 OpenCTI: User-Controlled ReDoS in JSON Ingestion Mapper

OpenCTI is an open source platform for managing cyber threat intelligence knowledge and observables. Prior to 7.260706.0, the JSON ingestion mapper's extractWithRegexp formula function compiled a user-supplied regular expression with the JavaScript RegExp engine in...

ROOT-APP-PYPI-CVE-2026-35526 CVE-2026-35526 in strawberry-graphql - Patched by Root

Root has patched CVE-2026-35526 in the strawberry-graphql package for Root:PyPI. Multiple fixed versions available...

ROOT-APP-PYPI-CVE-2026-35523 CVE-2026-35523 in strawberry-graphql - Patched by Root

Root has patched CVE-2026-35523 in the strawberry-graphql package for Root:PyPI. Multiple fixed versions available...

ROOT-APP-MAVEN-CVE-2026-41856 CVE-2026-41856 in org.springframework.graphql:spring-graphql - Patched by Root

Root has patched CVE-2026-41856 in the org.springframework.graphql:spring-graphql package for Root:Maven. Multiple fixed versions available...

ROOT-APP-MAVEN-CVE-2026-41700 CVE-2026-41700 in org.springframework.graphql:spring-graphql - Patched by Root

Root has patched CVE-2026-41700 in the org.springframework.graphql:spring-graphql package for Root:Maven. Multiple fixed versions available...

ROOT-APP-MAVEN-CVE-2026-41699 CVE-2026-41699 in org.springframework.graphql:spring-graphql - Patched by Root

Root has patched CVE-2026-41699 in the org.springframework.graphql:spring-graphql package for Root:Maven. Multiple fixed versions available...

CVE-2026-76821: Uncontrolled Resource Consumption

OpenCTI is an open source platform for managing cyber threat intelligence knowledge and observables. Prior to 7.260706.0, the JSON ingestion mapper's extractWithRegexp formula function compiled a user-supplied regular expression with the JavaScript RegExp engine in...

PT-2026-92468

OpenCTI is an open source platform for managing cyber threat intelligence knowledge and observables. Prior to 7.260701.0, the synchronizerFetch GraphQL query called fetchRemoteStreams after checking only that a remote stream URL used HTTP or HTTPS. The backend did not apply the ingestion deny lis...

PT-2026-99159

Summary @zereight/mcp-gitlab exposes GitLab to an LLM agent while relying on read-only mode, a project allow-list, and transport auth as its safety controls. Five defects defeat those controls. Under the MCP threat model, tool-call arguments/content can be shaped by untrusted input prompt injecti...