Company News
Introducing Arcjet's custom AI security models
Arcjet's prompt injection detection now runs on Arcjet Himalia 1, a custom fine-tuned model, with specialist models for coding and security agents.
Arcjet blog
Feed
More recent
Engineering
How we built AI agent behavior drift analysis with Cloudflare Clef
Company News
Introducing Arcjet for coding agents: security policies for Claude Code, Codex, Cursor, Muse Code, and Copilot
Company News
Introducing Agent Runtime Security: Observe, Enforce, Audit
Recent posts
Older notes and announcements from the Arcjet team.
- AIThe two speeds of AI agent runtime security
- ChangelogSDK releases, security briefing improvements, new activity view
- EngineeringRunning PII detection locally with the Rampart NER model
- ChangelogSDK performance releases and proxy detection
- EngineeringMaking Arcjet's Wasm bot detector smaller and faster
- Company NewsAnnouncing advanced bot signals to detect automation without CAPTCHAs
- EngineeringDesigning a CLI for AI agents
- ChangelogUnified skills, Python docs, email management
- EngineeringServing AI models with Open Inference Protocol APIs
- ChangelogArcjet CLI and team management
- EngineeringHow we defend MCP tool outputs from prompt injection
- Company NewsIntroducing Arcjet Guards: security inside the agent loop
- EngineeringBuilding a production MCP server in Go
- ChangelogAdvanced client signals, request filtering, build with agents
- EngineeringFrom devcontainers to VMs: parallel dev environments for AI agents
- ChangelogArcjet MCP server, coding agent plugins, weekly security briefing