yaad

A local memory engine for your terminal and your AI agents.

You forget the staging DB port. The curl flag that fixed that weird error. The deploy steps for this service. yaad saves it — and gives it back when you ask. Wait, not just you, your AI assistants can use it to store context across sessions!

Runs entirely local via Ollama. No cloud, no accounts, no API keys.

$ yaad add "staging db is postgres on port 5433"
$ yaad add "deploy checklist: run migrations, restart workers, clear cache"

$ yaad ask "what port is staging on?"
→ Staging DB is postgres on port 5433.

$ yaad ask "what's the deploy checklist?"
→ Run migrations, restart workers, then clear cache.

---

Install

Linux / macOS:

curl -fsSL https://yaad.knl.co.in/install.sh | bash

Go install:

go install github.com/kunalsin9h/yaad/cmd/yaad@latest

Pre-built binaries — GitHub Releases.

Then pull two Ollama models (one-time):

ollama pull mxbai-embed-large  # embeddings
ollama pull llama3.2:3b        # reasoning (or any chat model you prefer)

Requires Ollama running locally.

---

Reminders

yaad can remind you of things right in your terminal when they're due.

yaad add "submit PR for review" --remind "tomorrow 9am"
yaad add "book conference ticket" --remind "in 30 minutes"

See REMINDERS.md for setup (shell hook or systemd daemon).

---

Agent Integration

yaad ships as an Agent Skill — compatible with Claude Code, Cursor, Codex CLI, Gemini CLI, and any agent that supports the open skills standard.

npx skills add kunalsin9h/yaad -g

Once installed, your agent saves and recalls memory across sessions automatically.

---

More

• COMMANDS.md — full command reference (list, get, delete, config, all flags)
• REMINDERS.md — reminder setup guide
• CONFIG.md — configuration keys, models, notifier options
• SPEC.md — product specification and architecture

---

License

MIT © Kunal Singh