Granola Incident RunbookSAFE
Model-agnostic agent-skills platform with a harness-free canonical layer, verified adapters, and the ccpi package manager. Explore at tonsofskills.com.
Overview
Model-agnostic agent-skills platform with a harness-free canonical layer, verified adapters, and the ccpi package manager. Explore at tonsofskills.com.
4f83675ca38aOBSERVED · 2026-10-08Host compatibility
What the documentation claims. We have not run a compatibility test.
| Host | Status | Notes |
|---|---|---|
| claude-code | mentioned |
What it tells the agent
The instruction file, verbatim from the audited commit — this is the text the model reads, and the surface the audit's instruction layer examines. Quoted here so you can judge it without cloning anything.
--- name: granola-incident-runbook description: 'Incident response procedures for Granola meeting capture failures and outages. Use when meetings aren''t recording, transcription fails mid-meeting, integrations stop syncing, or the Granola service is down. Trigger: "granola incident", "granola outage", "granola down", "granola not recording", "granola emergency". ' allowed-tools: Read, Write, Edit, Bash(curl:*), Bash(pgrep:*), Bash(pkill:*), Bash(open:*) version: 1.13.0 license: MIT author: Jeremy Longshore <[email protected]> tags: - saas - granola - incident-response compatibility: Designed for Claude Code --- # Granola Incident Runbook ## Overview Standard operating procedures for Granola incidents — from individual recording failures to organization-wide outages. Covers triage, remediation, communication, escalation, and post-incident review. Designed for IT admins, team leads, and individual users. ## Prerequisites - Granola admin access (for org-level incidents) - Bookmark [status.granola.ai](https://status.granola.ai) for service status - Internal communication channel identified (#granola-support or similar) ## Instructions ### Step 1 — Triage: Assess Severity ```bash # Quick status check curl -s "https://status.granola.ai/api/v2/status.json" 2>/dev/null | python3 -c " import json, sys try: data = json.load(sys.stdin) indicator = data.get('status', {}).get('indicator', 'unknown') desc = data.get('status', {}).get('description', 'Unknown') print(f'Status: {indicator} — {desc}') except: print('Cannot reach status page — possible network issue') " || echo "Network error — check internet connection" ``` | Severity | Description | Response Time | Example | |----------|-------------|---------------|---------| | **P1 Critical** | Org-wide outage, data loss risk | Immediate | Granola service down, no one can record | | **P2 High** | Multiple users affected | < 1 hour | Recording fails for a team, sync broken | | **P3 Medium** | Single user issue | < 4 hours | One person's transcription stopped | | **P4 Low** | Minor issue, workaround exists | < 24 hours | UI glitch, slow processing | **Scope assessment questions:** 1. Is it just you, or are others affected too? → Ask in #granola-support 2. Is [status.granola.ai](https://status.granola.ai) showing an incident? → P1/P2 if yes 3. Was it working earlier today? → Recent change (OS update, permissions) likely 4. Which platform? (Zoom/Meet/Teams) → Platform-specific audio routing ### Step 2 — Remediation by Incident Type --- #### Incident: "Meeting Not Recording" **Severity:** P3 (single user) or P2 (team-wide) **Immediate actions:** 1. Click Granola menu bar icon > **Start Recording** (manual override) 2. If manual start fails: - Check that the meeting has audio playing - Verify Granola is running: `pgrep -l Granola` - Restart: right-click menu bar icon > **Restart Granola** **Root cause investigation:** - [ ] Calendar event has a video conferencing link (Zoom/Meet/Teams) - [ ] Calendar is connected in Settings > Calendar - [ ] Microphone permission granted - [ ] Screen & System Audio Recording permission granted (macOS) - [ ] Not running conflicting audio software (Loopback, BlackHole) **Backup:** Take manual notes. After fixing, re-record the next meeting. --- #### Incident: "Transcription Stops Mid-Meeting" **Severity:** P3 **Immediate actions:** 1. Check: is the computer awake? (Sleep kills transcription) 2. Right-click Granola icon > **Restart Granola** 3. Reopen your note — transcription may resume 4. If using Bluetooth: switch to wired audio or built-in speakers **Root cause:** Granola stops transcription after ~15 minutes of no detected audio. Bluetooth devices can cause intermittent dropouts. --- #### Incident: "Enhancement/Processing Failed" **Severity:** P3 **Immediate actions:** 1. Wait 15 minutes — long meetings take longer to process 2. Check internet connectivity 3. Check [status.granola.ai](https://status.granola.ai) for service issues 4. Restart Granola and reopen the note 5. Click **Enhance Notes** again **If still failing after 30 minutes:** The transcript was captured but enhancement may be queued. Submit support ticket with the meeting date/time. --- #### Incident: "Integration Not Syncing" (Slack/Notion/HubSpot) **Severity:** P3 **Immediate actions:** 1. Settings > Integrations > check the target integration status 2. Disconnect and reconnect the integration 3. Re-share the note manually 4. For Zapier: check Zap history at zapier.com for errors **Common causes:** | Integration | Likely Cause | Fix | |-------------|-------------|-----| | Slack | Bot removed from channel | `/invite @Granola` in the channel | | Notion | Database deleted | Reconnect (new database created) | | HubSpot | OAuth token expired | Reconnect in Settings | | Zapier | Connection expired | Re-authenticate Granola in Zapier | --- #### Incident: "Granola Service Outage" (P1) **Severity:** P1 Critical **Immediate actions:** 1. Confirm at [status.granola.ai](https://status.granola.ai) 2. Switch to **backup note-taking** immediately: - Open a text editor or Google Doc - Take manual notes for active meetings - Notes can be combined with future Granola captures manually 3. Communicate to your team via Slack: ``` :rotating_light: Granola is currently experiencing a service outage. Status: https://status.granola.ai Impact: Meeting recordings and AI enhancement are unavailable. Workaround: Take notes manually in Google Docs or your preferred editor. I'll update when service is restored. Next update: [time + 30 min] ``` 1. Subscribe to status updates at status.granola.ai 2. Resume normal operation when status returns to Operational ### Step 3 — Escalation Path ``` Level 1: User Self-Service → Restart Granola, check permissions, verify audio → Time: 5 minutes Level 2: IT Support / Team Admin → Run debug bundle (granola-debug-bundle) → Check org-
Trust audit
SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | PASS |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | PASS |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- none-observed
- Network
- none-observed
- Shell
- none-observed
- Dependencies
- pinned
- Secrets in source
- none-found
Findings (0)
No findings outside the package's declared scope.
Gates applied: no_behavioural_pass.
4f83675ca38afull audit observations/trust-audit/skill/jeremylongshore__granola-incident-runbook.json · Report an issue / request a re-scanAudit history
Every audit this skill has had.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-08 | 4f83675ca38a | SAFE | B | 89 | first audit |
Questions
What does the Granola Incident Runbook skill do?
Model-agnostic agent-skills platform with a harness-free canonical layer, verified adapters, and the ccpi package manager. Explore at tonsofskills.com.
Is Granola Incident Runbook safe to install?
The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean skill reads B.
What can Granola Incident Runbook access on my machine?
The audit observed no filesystem, network or shell use at all in its source.
Which assistants does Granola Incident Runbook work with?
Its documentation mentions claude-code. That is what the text claims, not a compatibility test we ran.
How current is this page?
The grade is for one exact copy of the source (4f83675ca38a), read on 2026-10-08. The repository is watched, and a new audit runs when it changes — this is the first audit.