EdgeSutra AI documentation
Evidence-first edge AI compatibility, quantization planning and benchmark auditing. Know what is measured, what is declared and what is unknown.
v0.2.182/86 tests passed0 runtime dependenciesApache-2.0Honest: device execution NOT_RUN
Executive summaryEdgeSutra AI 0.2.1 is an evidence-first toolkit that tells engineering teams what is proven, what is merely declared and what is unknown wheFunctional specification1. Any failed static check → BLOCKED (exit 1).Architecture user / CI / AI-IDE ──► edgesutra CLI ──► core engine ──► sealed JSON artifacts ──► MD / HTML rendererTechnical referenceedgesutra doctor|profile|versionUser guideNode.js 22 or newer. Nothing else: no API key, no network, no GPU.Security and privacyLeast privilege, read-only by default, untrusted-input handling, no secrets in artifacts, honest limits.Evidence model and rules1. Never write "runs on", "supported on" or "optimized for" a device without MEASUREDONDEVICE.Test reportMachine-generated by npm run proof and copied here by build.mjs. Raw logs are in [test-results/](test-results/test-run.tap).Operations and release<https://github.com/shahvatsalofficial/edgesutra-ai> · package edgesutra-ai · bins edgesutra, edgesutra-ai.Legal and governanceEngineering documentation, not legal advice.Roadmap and FAQIs this a new inference framework? No. It audits claims about existing ones.GlossaryIndependent assurance and evidenceAn independent-style review of what this project proves, how, and what it does not. Every figure below is regenerated by npm run proof.Sample audit reportReal HTML output of
edgesutra auditRaw test log (TAP)Unedited machine output