Sazabi
Sazabi is an AI-powered observability platform built to help engineering teams find and fix production issues faster, without leaving Slack.Sazabi brings logs, metrics, alerts, issue context, and AI analysis into the conversations where your team already works.Getting startedInstall the app, connect your Sazabi organization, set an active project, invite @Sazabi to your incident channels, and start asking questions. Visit app.sazabi.com to set up your account.Ask questions in plain EnglishMention @Sazabi in any channel or send it a direct message. Sazabi searches connected telemetry, identifies anomalies, and responds with explanations and visualizations inside the thread.Examples:- "Why is our API latency spiking?"
- "Show me errors from the payment service in the last hour"
- "What changed right before this incident started?"Attach related files in Slack, and Sazabi can analyze them as part of its response.Alerts that tell you what's happeningReceive issue notifications with severity, status, impact, root-cause analysis, and suggested remediation. Shared Sazabi issue and thread links unfurl as rich previews with actions to mute, unmute, resolve, ignore, or start a fix.Charts and tables, inlineSazabi can generate timeseries charts, log tables, and other visualizations as inline images, making production data readable even on mobile.
-
-
-
-
-
-
- "Use Sazabi MCP to query checkout errors from the last 30 minutes"
- "Use Sazabi MCP to check open critical issues"MCP access is scoped to the connected Slack workspace, linked Sazabi user, and that user's organization and project permissions.App HomeThe Sazabi App Home lets you view integration status, switch projects, browse recent AI threads, and open the full Sazabi dashboard.Slack plan requirementSazabi uses Slack's Agents & Assistants container for the Slack assistant and MCP experience. A paid Slack plan is required.AI accuracy noticeSazabi uses AI to generate answers, summaries, visualizations, and recommendations. AI output may be inaccurate or incomplete, so verify important information before taking action.Sazabi is an AI-native observability platform for engineering teams, built to replace traditional observability tools with a faster, Slack-native approach to production monitoring and incident response.
- "Show me errors from the payment service in the last hour"
- "What changed right before this incident started?"Attach related files in Slack, and Sazabi can analyze them as part of its response.Alerts that tell you what's happeningReceive issue notifications with severity, status, impact, root-cause analysis, and suggested remediation. Shared Sazabi issue and thread links unfurl as rich previews with actions to mute, unmute, resolve, ignore, or start a fix.Charts and tables, inlineSazabi can generate timeseries charts, log tables, and other visualizations as inline images, making production data readable even on mobile.
/sazabi slash commandsUse /sazabi to manage workspace configuration without leaving Slack:- /sazabi help - list available commands-
/sazabi projects list - view all available projects-
/sazabi projects use <id> - set the active project for your workspaceUse Sazabi through MCPSazabi provides a Model Context Protocol (MCP) server for Slackbot and other MCP-compatible AI clients. Once your workspace is connected and your Slack identity is linked, Slackbot can discover and run authorized Sazabi actions from Slack.At the MCP protocol level, Sazabi exposes:- search - discover available Sazabi operations, schemas, and examples.-
execute - run an authorized operation with validated input.Examples:- logs.query - search logs by service, severity, text, time range, and fields.-
issues.list / issues.search - find issues by status, severity, and name.-
threads.create / threads.get - start or reopen AI investigations.-
automations.list / automations.runs.list - review automation health and recent runs.Example Slackbot prompts:- "Use Sazabi MCP to list my projects"- "Use Sazabi MCP to query checkout errors from the last 30 minutes"
- "Use Sazabi MCP to check open critical issues"MCP access is scoped to the connected Slack workspace, linked Sazabi user, and that user's organization and project permissions.App HomeThe Sazabi App Home lets you view integration status, switch projects, browse recent AI threads, and open the full Sazabi dashboard.Slack plan requirementSazabi uses Slack's Agents & Assistants container for the Slack assistant and MCP experience. A paid Slack plan is required.AI accuracy noticeSazabi uses AI to generate answers, summaries, visualizations, and recommendations. AI output may be inaccurate or incomplete, so verify important information before taking action.Sazabi is an AI-native observability platform for engineering teams, built to replace traditional observability tools with a faster, Slack-native approach to production monitoring and incident response.
Sazabi will be able to view:
Sazabi will be able to do:
Review the details to better understand this app’s security practices. To learn more about assessing apps for your workspace visit our Help Center.
General
Developer
Sazabi
Company headquarters location
United States
Terms of service
Privacy & data governance
Data retention policy
Customer log and telemetry data is stored in ClickHouse and retained according to dataset-specific retention windows. Raw log data is currently retained for the active raw-log retention period, while aggregated log-volume data used for usage tracking and reconciliation is retained for 30 days. Account, organization, user, and integration metadata is stored in PostgreSQL/Supabase and retained for the duration of the customer relationship or applicable agreement, unless deletion is requested or retention is otherwise required by law.
Data archiving and removal policy
Sazabi uses soft deletion for account, organization, membership, and integration records in deletion and disconnect flows. App uninstall or token revocation soft-deletes or disconnects the relevant provider-specific workspace, user, installation, or connection records. Log and telemetry data is automatically purged after the applicable retention window. Customers may request permanent deletion by contacting legal@sazabi.ai; GDPR requests are handled within one month where applicable. Sazabi does not retain customer log and telemetry data beyond the active retention period except where required for backups, operational records, legal obligations, or separately governed service archives.
Data storage policy
Customer log and telemetry data are stored in ClickHouse in the region selected for the project at the time of creation. Supported production regions currently include us-west-2, us-east-1, us-east-2, and eu-central-1. Account and application metadata is stored in a US-hosted PostgreSQL/Supabase database. Sazabi encrypts data at rest and in transit.
App/service has sub-processors
yes
Guidelines for sub-processors
App/service uses large language models (LLM)
yes
LLM model(s) used
OpenAI, Anthropic Claude, OpenRouter, Vercel AI Gateway/Amazon Bedrock, and selected hosted model providers such as Baseten, depending on the product surface and configured model route.
LLM retention settings
Sazabi retains conversation/thread content and LLM metadata as service data. Raw LLM exchanges may be archived for debugging where configured. Provider retention is governed by the selected provider and route.
LLM data tenancy policy
Each LLM request is assembled within the authenticated customer workspace/project context. Prompts may include customer-authorized data from that workspace/project, including connected integrations, telemetry, thread history, attachments, and tool results
LLM data residency policy
LLM inference is performed by third-party AI infrastructure & routing providers. Unless a provider-specific regional route is explicitly configured, prompts and completions may be processed in the US or other regions where those subprocessors are located.
Certifications & compliance
Data deletion request procedure
Data deletion requests may be submitted to legal@sazabi.ai. Sazabi reviews and processes deletion requests within one month where required by applicable law, subject to legal, security, backup, and operational retention requirements. When a Slack workspace is disconnected or the app is uninstalled, Sazabi revokes or disconnects the relevant Slack integration and soft-deletes the provider-specific workspace and user records associated with that installation.
HIPAA compliant
yes
While this app may offer HIPAA compliance, Slack does not have a business associate agreement with any third-party application providers, including those in the Slack Marketplace, so you are responsible for validating the provider's compliance and executing an appropriate agreement before enabling.
Security
Date of latest pen test
2026-03-04
Executive summary is available to potential customers upon request
yes
Supports Single Sign On (SSO) with the following providers
Google OAuth / Google Workspace sign-in; email magic link
Supports Security Assertion Markup Language (SAML)
no
Has a dedicated security team
no
Contact for security issues
security@sazabi.ai
Has a vulnerability disclosure program
no
Has a bug bounty program
no
Requires third party authorization/connections
no
Uses token rotation
no