AIIA — AI skill for Claude Code
⌬ AIIA — the memory layer your AI tools should have.
How to install AIIA
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open ericlovo/AIIA and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What AIIA does
⌬ AIIA — the memory layer your AI tools should have. Persistent decisions, story-driven workflow, local-first LLM stack. Apache 2.0. Built at Aplora AI.
Alternatives in AI
README
⌬ AIIA — AI Information Architecture
A·I·I·A. A palindrome on purpose: the architecture mirrors the name.
Your memory on one side. Your AI tools on the other. AIIA is the surface in between — where what you decide becomes what they know, and what they notice becomes what you remember.
Without it, every session starts blank. Every decision gets re-explained. Every pattern gets re-derived. With it, one continuous thread runs between your thinking and the tools that act on it. Claude Code, Cursor, Copilot — pick your tool. The mirror works the same.
Open source. Apache 2.0. Local-first. Free model. Your data stays yours.
[](./CHANGELOG.md) [](./LICENSE) [](./pyproject.toml) [](#status)
Quick start
pip install aiia # PyPI install
aiia # first-run check + setup prompt
aiia next # surface your top story to work on now
aiia ask "what did I decide about lockfile policy?"
That's the daily loop. `aiia next` chooses, `aiia ask` recalls, `aiia memory add` captures.
If you don't want PyPI, clone and `pip install -e .[dev]`. AIIA needs Python 3.10+, ~5GB free for a local model, and either Ollama running on `localhost:11434` or an `ANTHROPIC_API_KEY` if you prefer cloud tiers.
What flows through the mirror
**From you, into your AI tools' working context:**
- Decisions you've made, with the why — recoverable, not buried in a chat log
- Patterns and conventions you've named — applied without re-explaining
- Stories in your backlog — surfaced when relevant, ranked by priority × freshness
- Session history — what you were doing when you stopped
**From your AI tools, back into your record:**
- Observations from running code — what changed, what
Related Skills
Singularmem
Singularmem is a local-first persistent memory layer for LLM-driven workflows built in Rust, SQLite, Typescrip
Sessionmem
Local-first MCP memory layer for AI coding assistants; persistent cross-session memory, stored entirely on you
Vibe Writing Open Source
A local-first AI writing workbench with Markdown editing, reviewed AI changes, research, memory and text skill
Somora Agent
Local-first gateway for personal AI agents with persistent memory across multiple LLM providers
AI Operator Kit
Battle-tested patterns from running AI agents in production daily: the CLAUDE.md that finishes work, a 5-layer
Recallium
Recallium is a local, self-hosted universal AI memory system providing a persistent knowledge layer for develo
Related Agents
Backend Eng
Use for the schema, contract, repository, service, and controller layers of Phase A, once a module is in scope
Path Prover
Produces live evidence that every layer a lane touched actually works — boots the local stack or targets stagi
Local Model Specialist
Evaluate on-device LLM integration, GGUF quantization levels, and KV cache memory constraints read-only.