GitHub Copilot can now operate desktop apps, not just code
The useful boundary change is that Copilot can now cross from code and terminals into ordinary desktop interfaces, with per-app approval and organisation-level controls.
Search
Search results for “MCP server”
Find published research by company, product, platform or technology.
Showing 41–60 of 188 dossiers
Ada gives customer-service agents sandboxed Python tools with controlled network access
Ada has added code tools that run a restricted Python subset inside agent conversations. They can transform API responses, perform deterministic calculations and call allowlisted domains, while MCP-authored changes can be staged and reviewed before promotion.
FrankenPHP 1.13.4 fixes seven security flaws, including unauthenticated worker-mode crash
Operators of FrankenPHP 1.13.x should upgrade promptly: 1.13.4 closes seven server flaws, including a worker-mode denial of service and request-isolation leaks.
Symfony Language Tools reaches 0.24 with richer Twig analysis, CI caching and Windows editor packages
The LSP is still pre-1.0, but shipped releases now cover wider Symfony frameworks and more developer environments, with improved caching and diagnostics.
GitLab patches a CVSS 10.0 CI/CD regex RCE — and five more high-severity flaws
CVE-2026-97229 carries GitLab's maximum 10.0 CVSS score. The 19.3 patch line requires 19.3.5, not the initially issued 19.3.4, because of an Ona integration regression.
Cloudflare WAF can now correlate malicious requests with origin responses and block the reply
Request-only WAF rules cannot always tell whether an exploit succeeded. Cloudflare now correlates requests with origin responses, adds managed response blocking and permits three custom response rules.
Google Analytics can now reject ghost traffic with hostname allowlists
Google Analytics hostname allowlists reduce some ghost-traffic cleanup, but server-side Measurement Protocol events bypass the filter. Teams must distinguish browser collection from Measurement Protocol and test empty-hostname events.
Tailscale MagicRoute beta changes how private networks select routers and service backends
Tailscale's new MagicRoute beta lets eligible organisations choose priority-based routes per client rather than regional grouping. It affects multi-region subnet access, service hosts and exit-node recommendations.
WordPress 7.2 will disable MySQL NO_BACKSLASH_ESCAPES for its own connections
WordPress 7.2 adds NO_BACKSLASH_ESCAPES to wpdb's incompatible SQL modes. It's a targeted fix for escaping failures, not a server-wide setting or blanket SQL injection cure.
DeepSeek opens Search and Fetch APIs at $0.75 per 1,000 calls, with a temporary agent-tool free window
Developers get another paid web retrieval option for coding agents and applications; the introductory harness exemption expires October 31.
Web DevelopmentCloud & InfrastructureCloud PlatformsDeployment & DevOpsFrameworksDeveloper ToolsActive dossier
Cloudflare opens Void: a Vite SDK, CLI and self-hosted deployment platform
Void SDK adds server routes and typed data access to Vite, Void CLI provisions Workers resources, and Void Platform provides a controlled multi-user deployment surface. The convenience comes with Cloudflare runtime dependencies and real deployment limits.
PostgreSQL JDBC 42.7.14 closes authentication and SQL-data leakage flaws
Two pgJDBC vulnerabilities affect different older driver ranges and only specific connection or binary-write configurations. Upgrade and verify the affected paths rather than treating this as a PostgreSQL server vulnerability.
Laravel AI SDK patches an SSRF path in its Vercel and AG-UI file adapters
The bug is a useful warning for AI application plumbing: turning a user-supplied URL into a model attachment also turns the application server into a network client unless the adapter enforces an outbound trust boundary.
The Events Calendar exposes 600,000+ WordPress sites to two comment-preview RCE chains
The dangerous detail is the delivery path: WordPress gives an unauthenticated commenter a moderation-preview URL for their own pending comment, and The Events Calendar can process attacker-controlled block markup from that preview before a moderator approves anything.
WordPress 7.1.1 closes Click2Shell’s forced-theme-install path to code execution
Click2Shell turns a theme-preview parsing bug into a supply-path problem: an attacker can force official catalog code onto a site without the administrator choosing Install, then potentially reach executable pre-activation theme code.
Google must open AdX and DFP to Prebid under six-year ad-tech remedies
Google must build Prebid integrations, let rival publisher ad servers receive real-time AdX bids, make publisher data portable and stop preferential AdWords bidding under a six-year court-supervised remedy.
GitHub has switched off SHA-1 HTTPS support — old Git and API clients can now fail
The deadline is no longer theoretical: browsers, Git HTTPS backends and API clients that still depend on SHA-1-era TLS algorithms can now lose connectivity to GitHub.com.
OpenRouter can now fail AI requests rather than let regulated data leave the US or EU
The important change is at the gateway boundary, not just inference placement. OpenRouter says prompts can now stay in-region from decryption through provider execution and supported server tools, while teams can enforce the rule per workspace, team or API key.
Super Forms’ 13,000-site WordPress flaw is being exploited at scale despite its smaller install base
This is a useful reminder that exploitation pressure does not scale neatly with plugin popularity: Wordfence says it has blocked more than 250,000 attempts against a plugin with a five-figure install base.
Artificial IntelligenceWeb DevelopmentBrowsers & Web StandardsInference & APIsDeveloper ToolsActive dossier
SwarmLLM splits a 27B model across browser tabs with WebGPU and WebRTC
SwarmLLM does not route whole prompts to separate machines; it pipelines one model across browser tabs. A MacBook and iPhone can jointly hold Qwen 3.8 27B even when neither device can hold the full 15GB quantized model alone, with no inference server in the loop.