Find published research by company, product, platform or technology.
Showing 1–20 of 70 dossiers
The change turns cache poisoning from mostly a workflow-design warning into an enforceable permission boundary. Teams can let untrusted jobs restore caches without writing them, prevent reusable workflows from escalating cache access and isolate jobs that only need to publish cache entries.
Edge Scripts no longer have to sit behind bunny.net’s automatic cache. New pre-cache hooks can execute before cache lookup, while a separate Cache API lets scripts read, write and delete regional cache entries under application control.
Cloudflare discloses cross-tenant GLM-5.3 Workers AI responses caused by a KV-cache rejection bug; SGLang v0.5.22 is recommended for disaggregated inference operators.
A detailed engineering case study links a 68% AI-review cost reduction to prompt-cache structure and effort tuning, with security-review trade-offs still unresolved.
Cloud CDN can now honor CDN-Cache-Control separately from browser-facing Cache-Control. That gives builders a standards-based way to set shared-cache behavior at the edge while preserving different client-side caching rules.
16.4 fills gaps in the new navigation/caching model and adds ensureStatic checks, while preserving a separate migration path for existing routes.
Inertia.js 3.9 flushes prefetch entries when history is cleared, preventing a previously visited protected page from reappearing from client cache after logout.
Astro 7.2’s experimental incremental-build mode attacks the page-generation phase rather than only bundling speed. Large static sites can opt routes into cache-aware reuse, but teams must choose correct cache keys and persist Astro’s cache directory in CI to benefit safely.
Umami v3.4.1 fixes a critical Next.js next/og RCE, image-optimisation SSRF, cache poisoning, an MCP OAuth credential leak and other dependency flaws. Self-hosted users should upgrade and rebuild or pull fresh Docker images; no migration is required.
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.
Chrome 155 no longer fires unload handlers by default. Sites relying on unload for saves, analytics or cleanup should migrate to visibilitychange, pagehide and resilient delivery paths.
The concrete API price and access limits close a key gap in the October launch: developers can cost Wonder through Model API but must not confuse maximum reasoning effort with Wonder Full.
Redis-compatible cache and queue costs have fallen across Render's 1GB-plus Key Value plans, with two new memory tiers and automatic repricing for existing customers.
Haiku 5.5 resets the economics of high-volume classification, extraction and agent sub-tasks, while Anthropic also cuts Sonnet 5.5 cache-read prices and introduces API credits for Max/Team subscribers.
The practical change is that debugging a Cloudflare-backed application no longer has to stop at the Worker boundary: one trace can follow security, cache, routing, Worker and origin handling, while logs and traces move toward one query and pricing model.
CLM-8B targets the same narrow decision layer as Jev, but with open weights, local deployment and a contrastive architecture that separates state and action representations. The headline speed and coding results are researcher-produced and need careful interpretation.
This was not a Firecracker escape or access to a live victim disk. It was a storage-isolation failure underneath the sandbox: researchers recovered foreign directory structures, database pages and complete SQLite databases from reused blocks, and Cloudflare had to fix allocation plus retire existing disks and cached snapshots.
Vercel Agent now works in Slack as well as the Vercel dashboard, combining logs, metrics, deployments and repository context with team conversation before proposing approved actions such as pull requests, rollbacks, configuration changes and cache purges.
Jev, CLM and GLiNER2.5-Decide made bounded software decisions look like a distinct model category. OpenAI is now validating the same architectural split with a Luna-powered API designed to answer finite questions rather than generate open-ended prose.
OpenAI's Ultrafast tier has moved beyond its GPT-5.6 preview: GPT-6.1 Sol access is broad and pricing is documented at 6× Standard.