メニュー
Query the user's browser-extracted profile: identity, accounts, tools, contacts, addresses, payments. Use when the user asks about their own info or you need personal context.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Use when the user says 'review memories', 'clean up the memory database', 'process unreviewed memories', 'memory cleanup', 'merge duplicate memories', or wants periodic LLM-powered post-ingestion review. Removes junk, merges duplicates, fixes miskeyed data, marks good entries as reviewed.
Set up ai-browser-profile for a new user. Installs via npm, creates Python venv, extracts browser data, and optionally enables semantic search. Use when: 'set up browser profile', 'install ai browser profile', 'configure browser profile'.
Analyze WhatsApp Web data — contacts, groups, social graph, AND decrypt actual message content via live browser interception. Use when: 'WhatsApp contacts', 'WhatsApp groups', 'WhatsApp analysis', 'who do I talk to', 'WhatsApp messages', 'decrypt WhatsApp', 'read WhatsApp messages', 'WhatsApp network', 'social graph', 'inner circle'.
Extract structured autofill data (names, emails, phones, addresses, companies) from Chromium browser 'Web Data' SQLite files. Use when: 'autofill data', 'browser addresses', 'saved addresses', 'autofill profiles', 'who is this person', 'extract contact info from browser', 'browser PII', 'form data'.
| name | ai-browser-profile |
| description | Query the user's browser-extracted profile: identity, accounts, tools, contacts, addresses, payments. Use when the user asks about their own info or you need personal context. |
Locally-extracted profile of the user built from their browser data (autofill, saved logins, history, bookmarks, WhatsApp, LinkedIn). Stored in ~/ai-browser-profile/memories.db. Nothing leaves the machine.
query_browser_profileUse this tool proactively whenever the user asks about themselves or you need personal context:
| User asks... | Query |
|---|---|
| "What's my email?" | query: "email address", tags: ["contact_info"] |
| "What accounts do I have?" | query: "saved accounts", tags: ["account"] |
| "What tools do I use?" | query: "tools and services", tags: ["tool"] |
| "Find contact X" | query: "X", tags: ["contact"] |
| "What's my address?" | query: "home address", tags: ["address"] |
| "What card do I use?" | query: "payment card", tags: ["payment"] |
| "Who am I?" / profile | query: "profile", tags: ["identity"] |
query_browser_profile(
query: string, // natural language query
tags?: string[] // optional: identity, contact_info, account, tool,
// address, payment, contact, work, knowledge
)
Returns ranked results from the local database. Results are self-ranking — frequently accessed ones surface automatically.
To get the complete user profile in one call:
query_browser_profile(query: "full profile")
Returns name, emails, phone, addresses, payment info, companies, top tools, accounts.
Requires browser data extraction during onboarding. If queries return no results, call extract_browser_profile to re-run the extraction — it uses the native Swift extractor built into the app (no external tools needed).