Slack Incident Broadcast Drafter
Incident updates that are clear under pressure.
Helps teams communicate incidents consistently without requiring operators to rewrite messy thread context into a clean broadcast.
Turns incident context and channel activity into structured Slack broadcasts for internal teams or customer-facing incident rooms.
How it runs
Multi-agent orchestration — here's the flow, step by step.
Call verslay_recall to retrieve prior incident templates, on-call rosters, and escalation channel configurations from memory. Then pull the last 50 messages from the configured incident channel to extract: incident title, time of detection, affected systems, initial impact description, current status, and the names of engineers actively responding. Identify the incident commander if mentioned. Store raw incident context as Phase 1 output.
slack channel operations agentConsume the affected systems and incident type from Phase 1. Call verslay_web_search and verslay_news_search to check for any known ongoing third-party service outages (e.g., AWS, Stripe, Cloudflare) that may be contributing to the incident. Call verslay_rss_fetch on the status pages of the affected infrastructure providers if their RSS feeds are configured. Determine whether the incident appears to be internal or external-dependency-caused. Pass attribution context to Phase 3.
web researcherIn parallel with web-researcher, scan Phase 1 Slack thread for sentiment signals: panic level (all-caps, exclamation frequency), confusion indicators (repeated 'what happened?', 'is this resolved?'), and clarity gaps (questions asked but not answered). Produce a communication health score and a list of open questions from the thread that need to be addressed in the broadcast.
sentiment analystConsume Phase 1 incident context, Phase 2 attribution context, and Phase 2 open questions. Synthesize the incident narrative: what happened, when, what is affected, what is the current status, who is responding, and what is the estimated resolution timeline. Identify the communication tier needed (Tier 1: all-hands Slack broadcast; Tier 2: stakeholder email; Tier 3: customer status page update). Apply incident communication best practices from verslay_recall.
knowledge synthesizerConsume Phase 3 synthesized incident narrative and communication tier. Draft broadcast messages in the appropriate format for each channel: (a) Slack broadcast — structured message with impact summary, current status, ETA, and on-call contact; (b) stakeholder email draft — formal incident notification with affected services, business impact, and next update time; (c) status page update text — concise public-facing description. Apply incident communication tone (factual, calm, action-oriented).
message writerConsume Phase 4 draft messages and Phase 2 open questions. Review drafts for completeness: confirm all open questions from Phase 2 are addressed or explicitly flagged as under investigation. Write a post-incident communication log: drafts sent, channels notified, open questions status. Call verslay_memorize to store the incident details, broadcast drafts, and template learnings in memory.
executive briefing writerReceive the finalized broadcast messages from executive-briefing-writer. Post the Slack broadcast to the configured all-hands and stakeholder channels. Confirm delivery and log send timestamp and message IDs.
distributorRequired Agents
7- slack-channel-operations-agent
- web-researcher
- sentiment-analyst
- knowledge-synthesizer
- message-writer
- executive-briefing-writer
- distributor
Connections
Required
What it does
- Channel context scan
- Incident status extraction
- Missing fact detection
- Broadcast draft
- Next-update reminder
Example prompt
Draft a Slack incident update from this channel. Include current status, customer impact, owner, ETA, next update time, and open questions.
Ready to deploy Slack Incident Broadcast Drafter?
Start free. One click, full agent orchestration.
Get Started Free →