# Capture Note

> Save a piece of information as a note, or look one up, using the "advanced" sample MCP server. Use when the user says "remember this", "save this as a note", "make a note that...", or asks "what notes do I have about X" / "find my note about X".

- Skill: `staffinityai/capture-note` (Agent Skill)
- Install (CLI): `npx skillmds@latest add staffinityai/capture-note`
- Raw SKILL.md: https://api.skillmd.com/api/skills/staffinityai/capture-note/raw
- Safety review: pending (external: skill-scanner PASS, skillspector PASS)
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: AI & ML
- Author: StaffinityAI (https://skillmd.com/u/staffinityai)
- Updated: 2026-09-22
- Page: https://skillmd.com/skills/staffinityai/capture-note

---


# Capture note

This skill orchestrates the `advanced` sample MCP server (defined in `.mcp.json`)
to store and retrieve notes. It exists to demonstrate a skill that drives MCP
tools — the prose here tells Claude *when* and *how* to call them.

The server exposes these MCP components (tool names are prefixed
`mcp__advanced__` when the server is connected):

- Tool `create_note(title, body)` — store a new note, returns it with an `id`.
- Tool `search_notes(query)` — return notes matching a substring.
- Resource `notes://all` — every stored note (static resource).
- Resource `note://{note_id}` — one note by id (resource template).

## To save a note

1. Derive a short `title` and a `body` from what the user asked you to remember.
   If they didn't separate the two, use the first line / key phrase as the title
   and the full text as the body.
2. Call `create_note` with them.
3. Confirm back to the user with the returned note `id` and title.

## To find a note

1. Call `search_notes` with the user's keywords. If they ask for "all notes",
   read the `notes://all` resource instead.
2. Summarize the matches (id + title). If nothing matches, say so.

Note: the sample server keeps notes in memory and resets when it restarts, so
treat stored notes as ephemeral.