Radar / AI security / Perplexity SPACE
Perplexity SPACE — open-source Firecracker microVM sandbox for AI agents
Perplexity open-sourced SPACE (Sandboxed Platform for Agentic Code Execution), the Rust platform behind every Perplexity Computer task, under Apache 2.0.
Why it matters
Each agent task gets its own Firecracker microVM with pause/resume/snapshots, credentials kept outside the guest, and an E2B compatibility wrapper — production-tested inside Perplexity Computer and red-teamed by nine frontier models, now Apache 2.0.
What you could build with it
A team building browser or coding agents could self-host SPACE as the execution layer: every agent task runs in an isolated microVM with snapshots and egress controls, so a compromised agent cannot reach host secrets or the wider network.
Does it hold up?
Early but grounded — production-tested inside Perplexity Computer (millions of sandboxes per day) and red-teamed by nine frontier models with no VM escapes; self-hosting needs Linux x86_64 with KVM/btrfs/Docker, no macOS.
Built with Perplexity SPACE
- TokenPost: Perplexity open-sources SPACE sandbox for AI agents with root accesstokenpost · Covers SPACE's per-task microVMs, host-side credential handling, gateway-routed traffic, and the 185ms-to-60ms latency drop.
- Forkast: Perplexity ships its search agent's sandbox as open sourceforkast · Analysis of SPACE's E2B compatibility wrapper and CoreWeave/Akamai/Crusoe cloud partnerships for hosted SPACE.
- perplexityai/SPACEgithub · 1,602 ★ · Official Apache 2.0 Rust implementation: Firecracker microVMs per task, HTTP API, pause/resume/snapshots, E2B compatibility wrapper.
Learn more
First spotted on article: source.
More AI security
iFixAi: independent auditing of AI agentsIndependent AI-agent auditor: 250+ inspections across 69 misalignment categories (red teaming + operational assurance)…security · JEV 0.76Cloudflare security-audit-skill: six-phase agent-driven security audits with independently verified findingsOpen-source MIT coding-agent skill that runs a codebase through six audit phases — reconnaissance, coverage-led…security · JEV 0.74OpenAI, Anthropic and Google DeepMind jointly unveil cyber-focused safety models and safeguardsThe three rival labs disclosed weeks of behind-the-scenes coordination and jointly released a set of cyber-focused AI…security · JEV 0.73OpenAI textGrain watermarking for ChatGPT and Codex in the EUOpenAI will roll out its invisible textGrain watermarking system to ChatGPT/Codex users in the EU over the coming…security · JEV 0.65Cloudflare security-audit-skill: multi-phase security audits for coding agentsA coding-agent skill that runs multi-phase security audits with independently verified, machine-checkable findings.security · JEV 0.63LiveNerf: a pre-registered benchmark for post-release model driftOpen-source, pre-registered 30-day benchmark that detects whether Claude Opus 5.5 quietly gets worse after launch.security · JEV 0.61
Get the week's best AI launches, plus 3 ideas worth building
One email every Saturday. Ranked by traction, not hype. Free.