Live chat punishes hesitation. The customer is waiting in real time while the agent digs for the right doc or remembers how the last person phrased this. Slow replies feel worse in chat than anywhere else.
sonnet1 weekIntercomOpenAIAlgolia
Claude
98ROI
72Scale
$5k94Saved
ROI for
README.md
Why this subagent
Live chat punishes hesitation. The customer is waiting in real time while the agent digs for the right doc or remembers how the last person phrased this. Slow replies feel worse in chat than anywhere else.
This copilot suggests the next reply right inside the chat window, grounded in your docs and past tickets, ready to send or edit in one click. It learns from the edits agents make, so the suggestions get closer to your team's voice over time. Agents answer faster without copy-pasting from a wiki in another tab.
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.
Search docs, past tickets, and macros for the most likely answer.
pending
Suggest a reply to the agent inside the chat window for one-click send or edit.
pending
Record which suggestions were accepted, edited, or rejected to sharpen the next ones.
pending
Sample output
json
// Sample output
// (generated when the pipeline finishes)
During a live chat, suggest the next agent reply grounded in retrieved context, and return it as editable draft text.
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.