Research Deep-Dive Index

The pages under docs/research/ are deep-dive reports informing capOS design decisions. Proposals and design notes cite them as grounding for capability model, IPC, scheduling, networking, error handling, runtime, agent, and prior-art choices. The Capability-Based and Microkernel Operating Systems survey records the cross-system design consequences pulled from this body of research; the entries below give the full alphabetical listing of individual reports for direct discovery.

Start here:

• Research: Capability-Based and Microkernel Operating Systems – Cross-system survey synthesizing the design consequences for capOS (capability table, IPC, memory, scheduling, persistence, VFS, resource accounting, language support).

Individual reports:

• Research: Browser Engines, Document Engines, and Agent Browsers – Browser engine portability, cap-native document-engine options, and agent-browser patterns for capOS browser capabilities.
• Cap’n Proto Error Handling: Research Notes – Prior-art on capnp-rpc error semantics.
• Cloud DMA Provider Evidence Inventory – Official AWS/Azure/GCP device-surface facts, evidence-matrix schema, live guest-probe checklist, and fail-closed classification rules for the cloud DMA backend decision.
• Research: Cloudflare, Cap’n Proto, Workers RPC, and Cap’n Web – Cloudflare Workers, workerd, Durable Objects, Workers RPC, Cap’n Web, and Cloudflare’s production use of Cap’n Proto/KJ.
• Completion Rings And Threaded Runtimes – Io_uring-style transports under threaded runtimes.
• Crash Recovery and Supervision – Prior-art survey of supervision trees, restart budgets, and death-observation semantics across Erlang/OTP, systemd, Kubernetes, Fuchsia, seL4, and Genode.
• Debug, Trace, and Profiling Authority – Prior-art survey of debug/trace/profile authority models (GDB RSP, Linux ptrace/Yama, perf/eBPF, Fuchsia, seL4, Genode) for the Debug and Trace proposal.
• DMA User-Space Driver Isolation – DMA, user-space driver, vIOMMU, and no-IOMMU bounce-buffer design consequences for capOS device authority.
• Research: EROS, CapROS, and Coyotos – Persistent capability-system lineage.
• Research: Future Scheduler Architecture – Survey of modern scheduler algorithms and architectures for capOS scheduler evolution.
• Research: Game Mechanics Prior Art – Grounded mechanics research for Aurelian Frontier seasonal play, markets, construction, and tactical combat.
• Genode OS Framework: Research Report for capOS – Componentized OS framework.
• Research: Hosted Agent Harnesses – OpenClaw-like harnesses, swarms, memory/wiki systems, and agent orchestration research for capOS-hosted agents.
• Research: HPC Parallel Patterns – HPC benchmark and programming-model grounding for generic parallel processing patterns.
• IOMMU Remapping Grounding – Primary-source grounding for future Intel VT-d, AMD-Vi, and QEMU IOMMU remapping work.
• IX-on-capOS Hosting Research – IX as a package corpus, content-addressed build/store model, and a capability-native build-service surface for capOS.
• Research: Linux Sandboxes And Virtualization For Workloads – Linux sandbox, container, gVisor, KVM, microVM, and CPU-isolation prior art for generic Linux workload execution.
• LLVM Target Customization for capOS – Requirements for a custom LLVM target triple.
• Research: Multimedia Pipeline Latency – Survey of PipeWire and JACK design lessons for the capOS multimedia graph.
• Research: NO_HZ, SQPOLL, and Realtime Scheduling – Linux NO_HZ, io_uring SQPOLL, CPU isolation, PREEMPT_RT, SCHED_DEADLINE, and seL4 MCS grounding for capOS timer and realtime design.
• OS Error Handling in Capability Systems: Research Notes – Cross-OS error-model comparison.
• Research: Out-of-Kernel Scheduling – Userspace scheduling prior art.
• Research: Paperclips Clean-Room Functional Spec – Clean-room functional mechanics summary and capOS Paperclips improvement candidates.
• Pingora Architecture and Philosophy: Research Report for capOS – Proxy/server framework as a userspace runtime case study.
• Research: Plan 9 from Bell Labs and Inferno OS – Namespace-oriented systems.
• Research: Realtime Multimodal Agent APIs – Provider API survey for realtime native-audio, multimodal, tool-using agents and their consequences for capOS agent surfaces.
• Research: Robotics Realtime Control – Robotics realtime-control practice and the consequences for using capOS as a robot brain.
• Research: Scientific Agent-Lab Software Stack – Scientific computing, solver, proof-assistant, notebook, and reproducible-package prior art for a capOS-hosted LLM research lab.
• seL4 Deep Dive: Lessons for capOS – Microkernel and capability reference.
• seL4 HAMR: Model-Based High-Assurance Engineering – Evaluation of HAMR (AADL/Slang/CAmkES) versus the capOS Cap’n Proto schema-as-contract model.
• Small Open-Weights LLM Survey for the capOS Agent-Shell – Model candidates for the on-ISO local LLM.
• Research: Spritely, OCapN, and CapTP – Spritely, OCapN, CapTP, netlayers, locators, Syrup, promise pipelining, handoffs, and capability-network lessons for capOS.
• Research: Time and Clock Authority – Prior-art survey of OS clock IDs, NTP/PTP discipline, leap-second handling, time namespaces, and Fuchsia UTC clock objects for the capOS time and clock design.
• x2APIC and APIC Virtualization – Interrupt routing on modern x86.
• Fuchsia Zircon Kernel: Research Report for capOS – Handle-based OS reference.