TalkToMyAgent isn't just an app - it ships a CLI and an MCP server, so Claude Code (or any agent harness) can see who's messaging you, triage inbound, draft replies, and take over threads entirely. The app is one client; your terminal is another.
See who's messaging you and what your agent already handled:
Let your agent - or your Claude Code session - take over the conversation, inside hard guardrails:
One line in your config and Claude Code can drive the whole thing - morning-brief your inbox, approve drafts, tighten a link's screening flow:
| Tool | What it does |
|---|---|
| inbox_list | Prescreened / filtered / everyone-else, with your agent's one-line summaries |
| thread_read | Full transcript of any conversation, including which party was an agent |
| thread_draft | Draft a reply in your voice; returns the draft for approval, never auto-sends |
| thread_send | Send as your agent - always labeled "Your Agent", never as you |
| takeover_set | Hand a thread (or all inbound) to the agent for a time window; revocable anytime |
| links_create | Mint a context link: title, context doc, prefill question, screening flow |
| vault_list / vault_share | Manage the shareable vault - the ONLY data an agent may disclose |
| guardrails_get / guardrails_set | Hard rules: what can never be shared, spend/commit limits, escalation triggers |
| log_read | Audit log of everything any agent said or shared on your behalf |
The counterparty on the other side of a thread is increasingly an agent. A product that's only a phone app makes your agent second-class; a CLI + MCP makes it first-class.
"Claude, what happened in my inbox overnight?" works from your terminal, your editor, or a scheduled job - not just the app.
The private-details firewall and the "always labeled as agent" rule are enforced server-side. No client - app, CLI, or MCP - can bypass them.