Some teams cannot send customer mail through a cloud API for privacy or compliance reasons. This auto-responder runs entirely on your own hardware, reading the inbox and drafting replies with a local model so no message body ever leaves the building.
sonnet5 daysIMAPSMTPOllamaLocal LLM
Claude
89ROI
83Scale
$4.7k97Saved
ROI for
README.md
Why this subagent
Some teams cannot send customer mail through a cloud API for privacy or compliance reasons. This auto-responder runs entirely on your own hardware, reading the inbox and drafting replies with a local model so no message body ever leaves the building.
It classifies each email first, answers only the safe ones, and escalates anything it is unsure about to a person. You get same-day replies on common questions without shipping data to a third party.
How it runs
Used at step 01 to kick off the pipeline.
Write
Used at step 01 to kick off the pipeline.
WebFetch
Used at step 01 to kick off the pipeline.
WebSearch
Used at step 01 to kick off the pipeline.
Run the local model through Ollama to draft a reply only for messages marked answerable.
pending
Apply sender rules so VIP addresses and unknown domains are skipped and routed to a human.
pending
Send the reply over SMTP and record the message id in SQLite to avoid double answers.
pending
Sample output
json
// Sample output
// (generated when the pipeline finishes)
Reply to this email using only the local knowledge provided; if unsure, return the single token ESCALATE.
Unlock the rest
The full agent definition, install snippet, and starter task are gated for community members.
Members get the full `.md` agent file, the npm / pnpm install one-liners, a starter prompt that we've tuned against real runs, and the open-source repo when this automation ships there. One email, magic link, done.