Skip to content
View norika1207-lab's full-sized avatar
25 followers · 0 following

Achievements

Achievement: Starstruck

Achievements

Achievement: Starstruck

Block or report norika1207-lab

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
norika1207-lab/README.md

Norika Oda

I work on AI agent systems: durable memory, agent observability, MCP/tool governance, and self-improving AI companions.

My current focus is the gap between impressive agent demos and agent systems that can be trusted in production. I care about the operational layer: what an agent remembers, what it can prove, how a human can inspect its work, and how teams recover when long-running automation goes wrong.

Current Research Threads

  • Agent memory as auditable operational evidence, not only larger context or retrieval.
  • Observability for agent state transitions, tool calls, memory writes, and rollback.
  • MCP and tool governance: permission boundaries, execution records, and runtime denial states.
  • Coding-agent reliability across long sessions, compaction, checkpoints, and remote starts.
  • Multi-agent trust: provenance, backpressure, budget controls, and handoff integrity.

Public Writing

What I Am Looking For

I am interested in practical collaborations around production agent memory, trace design, runtime evidence, and governance primitives for autonomous or semi-autonomous AI systems.

Author: Norika Oda (織田紀香) · https://norika.tw

Popular repositories Loading

  1. mercury-mcp mercury-mcp Public

    Cross-architecture LLM internal observation database (23 models, 13 architecture families). Exposed as MCP tools for any AI coding agent.

    Python 43 14

  2. Code-Duo Code-Duo Public

    Two AI coding agents — Claude & Codex — in one window. Stream every step, cross-review, watch the cost, and catch the AI that claims it did the work but didn't. Subscription-driven, no API.

    Python 38 8

  3. afu-brain afu-brain Public

    OpenClaw-compatible MASL safety gate with public RAG packs for memory-aware AI agents

    Python 23 4

  4. llm-neuron-atlas llm-neuron-atlas Public

    3D explorable atlas of every neuron in a transformer LLM. Click neurons, trace residual highways, compare cross-architecture conservation. Live demo: https://charenix.com/qwen3b-atlas

    HTML 20 5

  5. AI-Lies-Monitor AI-Lies-Monitor Public

    Python 10 3

  6. alfred-system alfred-system Public

    Personal AI agent system: background memory + parallel workers + Afu Brain safety gate

    HTML 5 2