S

Slack Workspace Agentic Workflow

Respond to Message — Slack Workspace Agentic Workflow

Generates a contextual response to a Slack message. Pure utility — the response tone and content are shaped by the injected role context.

Available free v1.0.0 Browser
$ sidebutton install slack
Download ZIP

A response-generation primitive for Slack. Given a source message and any supplied thread context, the workflow drafts an appropriate reply whose tone and content are shaped by whatever role context is injected at call time — an SE agent produces a technically precise answer, a support agent produces a customer-friendly one, and so on.

The workflow does not post the reply; it only produces the draft. Pair it with the browser-post-message primitive when actual sending is intended, or return the draft to the operator for manual review.

Steps

  1. 1.
    llm generate
    prompt
    |
    as
    response
    llm.generate
  2. 2.
    control stop
    message
    DRAFT_RESPONSE: {{response}}
    control.stop

Workflow definition

schema_version: 1
version: "1.0.0"
last_verified: "2026-02-08"
id: slack_respond
title: "Respond to Message"
description: "Generates a contextual response to a Slack message. Pure utility — the response tone and content are shaped by the injected role context."
overview: |
  A response-generation primitive for Slack. Given a source message and any supplied thread context, the workflow drafts an appropriate reply whose tone and content are shaped by whatever role context is injected at call time — an SE agent produces a technically precise answer, a support agent produces a customer-friendly one, and so on.

  The workflow does not post the reply; it only produces the draft. Pair it with the browser-post-message primitive when actual sending is intended, or return the draft to the operator for manual review.

category:
  level: primitive
  domain: ops
  reusable: true
params:
  message:
    type: string
    description: "The message to respond to"
  reporter:
    type: string
    description: "Who sent the message"
    default: "User"
  channel:
    type: string
    description: "Source channel"
    default: "general"
steps:
  - type: llm.generate
    prompt: |
      Write a brief, helpful response (2-3 sentences) to this message from {{reporter}} in #{{channel}}:

      "{{message}}"

      Respond naturally based on who you are and what you can help with.
    as: response

  - type: control.stop
    message: "DRAFT_RESPONSE: {{response}}"