> ## Documentation Index > Fetch the complete documentation index at: https://docs.sammylabs.com/llms.txt > Use this file to discover all available pages before exploring further. # Slack Integration > Secure, privacy-first Slack integration for policy notifications, JIRA ticket creation, and knowledge ingestion Sammy Labs provides a comprehensive Slack integration that transforms Slack workspaces into intelligent knowledge management and policy tracking hubs. This integration delivers three core features that work together to create a seamless workflow where policies are discovered, tracked, analyzed, and actionedβ€”all within Slack. ## Security First: Built with Privacy by Design **SAMMY on Slack follows the principle of least privilege** and is designed to only respond when explicitly invoked (e.g., an @mention in a thread). It does not inject itself into random public channels or send unsolicited messages. **From the app manifest, SAMMY is granted access only to:** `app_mentions:read` - Detect when users explicitly mention @SAMMY `chat:write` - Send responses and notifications to threads `files:read` - Access files in channels where SAMMY is added `channels:history` - Read messages only in public channels SAMMY is explicitly added to While Slack permissions allow SAMMY to read messages within channels it is added to, **SAMMY's application logic only processes and responds to user-triggered interactions** in channels where users have chosen to involve the bot. ### What This Means for Your Organization SAMMY never monitors or logs conversations in the background Processing only occurs when you @mention SAMMY or click an action button You control exactly when and where SAMMY operates *** ## Core Capabilities Automated policy updates delivered to Slack channels with rich formatting and actionable buttons One-click JIRA ticket creation from policy notifications with automated analysis Thread-based knowledge extraction and storage via @sammy mentions *** ## Feature 1: Policy Notifications to Slack ### Overview Policy notifications automatically deliver policy updates to configured Slack channels when new policies are detected or updated. These notifications include rich formatting, actionable buttons, and country-specific team mentions. ### User Experience When a new policy is detected, a formatted notification appears in your designated Slack channel: ``` πŸ“‹ POLICY UPDATE πŸ‡¬πŸ‡§ United Kingdom employment_law *New UK Employment Policy: Remote Work Guidelines* βœ… Status: Active βœ“ Verified: High Confidence πŸ“ Summary: The new UK Employment Policy establishes comprehensive guidelines for remote work arrangements, including eligibility criteria, equipment requirements, and performance expectations... πŸ’Ό Impact: This policy affects all UK-based employees working remotely. Key changes include mandatory equipment audits, updated performance review cycles... πŸ”— Sources: β€’ https://example.com/policy/remote-work-uk β€’ https://internal.wiki/remote-work-guidelines [Create JIRA Ticket] [View in Dashboard] ``` ### Features Modern Slack Block Kit components with emojis, visual indicators, and hyperlinks Long content automatically truncated with "See more..." links for readability Automatic tagging of country-specific teams based on policy jurisdiction Create JIRA tickets or view policies in dashboard with single click ### How It Works New or updated policies are automatically detected by the system Rich, formatted notification sent to your designated Slack channel Country-specific teams automatically mentioned for relevant policies Click "Create JIRA Ticket" or "View in Dashboard" to take immediate action The notification channel can be configured in your Sammy Labs dashboard settings. ### Use Cases Get notified immediately when new policies are detected, with full context for compliance review and action. Track policy changes with complete source documentation and verification status for legal analysis. Understand impact of policy changes on workflows and operational processes with automated summaries. Country-specific teams automatically tagged for relevant policy updates in their jurisdiction. *** ## Feature 2: JIRA Ticket Creation from Slack ### Overview Create JIRA tickets directly from Slack policy notifications with a single click. The system analyzes the policy in context, creates a JIRA ticket with analysis results, and sends confirmation back to Slack. ### User Experience User clicks "Create JIRA Ticket" button on policy notification Modal opens with pre-filled policy data (summary, description, priority, custom fields) User confirms settings and submits; receives "⏳ Processing policy analysis..." message System ingests thread, analyzes policy, and creates JIRA ticket with research Slack thread receives "βœ… A Jira ticket has been created: \[TICKET-URL]" message ### What Happens Next After you submit the JIRA ticket request: The system analyzes the policy in context to gather relevant research and insights A JIRA ticket is automatically created with the analysis results attached You receive a confirmation message in the Slack thread with a link to the JIRA ticket The entire process typically completes in under a minute. You'll receive status updates in the Slack thread so you know exactly what's happening. *** ## Feature 3: Knowledge Ingestion via @sammy Mentions ### Overview Mention @sammy in any Slack thread to ingest the entire thread as knowledge. The system processes text, videos, and attachments, extracts structured information, and stores it in your organization's knowledge layer. ### User Experience Simply mention the bot in any thread: ``` User: @sammy process this thread Bot: Got it! Your thread has been ingested to your knowledge layer. ``` **Behind the scenes:** * Complete thread is fetched from Slack * Videos are downloaded and analyzed * Text content is processed and structured * Memories are extracted and stored * Content becomes searchable in knowledge base ### How It Works Simply @mention SAMMY in any thread you want to capture SAMMY confirms immediately: "Got it! Your thread has been ingested..." All thread messages, videos, and attachments are processed in the background Extracted knowledge becomes searchable in your organization's knowledge base ### Video Processing When videos are detected in threads, the system performs advanced AI analysis: Videos are analyzed using Vertex AI with a specialized documentation extraction prompt that identifies: * Product features and how-to guides * Technical documentation (APIs, SDKs) * Workflows and processes * Configuration and setup instructions * Business information (pricing, SLAs) * Policies and guidelines * Troubleshooting guides **Video Processing Steps:** 1. **Detection** - Videos in the thread are automatically identified 2. **Secure Processing** - Videos are processed in secure, encrypted storage 3. **AI Analysis** - Advanced AI extracts structured documentation from video content 4. **Cleanup** - Temporary files are immediately deleted after processing 5. **Indexing** - Extracted content becomes searchable in your knowledge base ### Content Types Supported All thread messages with author attribution and timestamps MP4, MOV, and other formats with AI-powered content extraction Files and documents shared in thread URLs referenced in messages with context ### What Gets Captured **Content Stored:** * All thread messages with author attribution * Video transcripts and extracted insights * File attachments and shared documents * Links and references from the conversation * Full conversation context and threading **Privacy Options:** * **Internal** - Accessible only within your organization * **Public** - Available in your organization's public knowledge base You control the privacy settings when ingesting content. Internal memories remain completely private to your organization. ### Use Cases @sammy after a meeting to capture discussion, decisions, and action items automatically @sammy on demo threads to extract feature documentation from videos and discussions @sammy on training threads to create searchable knowledge base from educational content @sammy on policy discussion threads to preserve context and decisions @sammy on technical threads to build documentation from troubleshooting discussions ### Example ``` User: @sammy process this thread Thread Content: β”œβ”€ Message 1: "Here's how to enable 2FA..." β”œβ”€ Message 2: [Video: Product demo] └─ Message 3: "The API endpoint is /api/v2/auth/2fa" Bot: Got it! Your thread has been ingested to your knowledge layer. What Happens: β”œβ”€ All 3 messages captured with full context β”œβ”€ Video analyzed and documentation extracted └─ Everything becomes searchable in your knowledge base ``` *** ## Security & Data Protection ### How We Keep Your Data Safe Our integration implements enterprise-grade security to ensure your data remains protected and isolated: **Every Request is Verified** * All Slack requests are cryptographically verified to ensure authenticity * Timestamp validation prevents replay attacks * Only legitimate requests from Slack are processed **Your Credentials Stay Secure** * All workspace tokens are encrypted at rest * Credentials never exposed in logs or responses * Secure vault storage with enterprise-grade encryption **Complete Organization Isolation** * Your data is completely separated from other organizations * Database-level security prevents cross-organization access * Zero possibility of data leakage between organizations **Encrypted Transmission** * All communications use HTTPS/TLS encryption * End-to-end security for all data in transit * Industry-standard security protocols ### Minimal OAuth Permissions The Sammy Labs Slack integration follows the **principle of least privilege**, requesting only the absolute minimum permissions necessary for core functionality. We deliberately exclude permissions for private channels, DMs, and workspace-wide access. **Complete Slack App Manifest Scopes:** ```json theme={null} { "scopes": { "bot": [ "app_mentions:read", // Only detect @SAMMY mentions "chat:write", // Send responses in threads "files:read", // Access shared files "channels:history" // Read public channel messages (only where invited) ] } } ``` **What SAMMY Cannot Access:** * Private channels (unless explicitly invited) * Direct messages or group DMs * Channels it has not been added to * Workspace user lists or admin functions * Ability to delete or modify messages ### Exactly What SAMMY Can Access **What SAMMY will be able to view:** * **Messages** in public channels where SAMMY has been explicitly added * **@mentions** of SAMMY in conversations * **Files** shared in channels and conversations SAMMY has access to SAMMY can only access content in channels where it has been explicitly invited. It cannot see content in channels it hasn't been added to. **What SAMMY will be able to do:** * **Send messages** as @SAMMY to post analysis results and confirmations * **Reply to threads** where it was mentioned or where tickets were created * **Post notifications** for policy updates and JIRA ticket confirmations SAMMY only processes content when explicitly mentioned or invoked. It never monitors conversations in the background. ### Privacy Principles SAMMY only processes content when explicitly invoked by users via @mentions The bot never passively monitors conversations or logs content without explicit request All processing requires explicit user action (button click or @mention) Only permissions essential for functionality are requested ### Key Security Guarantees SAMMY never automatically joins channels. It must be explicitly invited using `/invite @SAMMY` SAMMY only posts when responding to @mentions or sending requested notifications Application logic only processes content when users explicitly invoke SAMMY All SAMMY actions are visible in thread replies and audit logs **Critical Privacy Note:** While Slack's `channels:history` permission technically allows SAMMY to read messages in channels it's added to, SAMMY's application code is designed to **only process and respond to explicit user interactions** (@mentions, button clicks). SAMMY does not passively monitor, log, or process conversations in the background. ### Data Handling & Privacy All your data is encrypted at rest using enterprise-grade encryption. Credentials and tokens are stored in secure vaults. All communications between Slack and Sammy Labs use HTTPS/TLS encryption to protect data in transit. Your organization's data is completely isolated. No possibility of access by other organizations. Files like videos are processed securely and immediately deleted from temporary storage after processing. *** ## Installation & Setup ### Getting Started Installation requires Slack workspace administrator permissions and an active Sammy Labs organization account. Navigate to your Sammy Labs dashboard, go to Integrations, and click "Add Slack Integration" Review the requested permissions on Slack's authorization page and approve the integration The bot is automatically configured with your organization's settings Add SAMMY to specific channels using `/invite @SAMMY` where you want to use the features You're ready to go! @mention SAMMY or receive policy notifications immediately ### Configuring Channels **Set Your Notification Channel** Configure which Slack channel receives policy notifications in your Sammy Labs dashboard under Settings > Integrations > Slack > Notification Channel. Choose a channel dedicated to compliance or policy updates for best organization. **Connect Your JIRA Account** To enable JIRA ticket creation from Slack, connect your JIRA account in your Sammy Labs dashboard under Settings > Integrations > JIRA. You'll need: * Your JIRA workspace URL * An API token with appropriate permissions JIRA integration is optional. Policy notifications work without it. **Control Where SAMMY Works** SAMMY only works in channels where it's been explicitly invited. To add SAMMY to a channel: 1. Open the channel in Slack 2. Type `/invite @SAMMY` 3. Press Enter To remove SAMMY from a channel, simply remove it like any other user. *** ## Frequently Asked Questions **Can SAMMY access all our Slack messages?** No. The integration only processes messages in threads where it's explicitly mentioned by users via @mentions. It never passively monitors conversations. **How is our data isolated from other organizations?** Complete isolation through database-level Row Level Security, separate encryption keys, organization-specific bot tokens, and multi-tenant architecture. **Can we control what data is processed?** Yes. Processing only occurs when users explicitly mention @sammy or click action buttons. No background monitoring or automatic processing occurs. **Can we revoke access?** Yes. Access can be revoked instantly through either Slack's app management interface or your Sammy Labs dashboard, immediately stopping all bot functionality. **What happens if we uninstall the integration?** Extracted memories remain in your knowledge base for continued access. Raw Slack data is never stored, and bot tokens are automatically invalidated. **How do we manage user permissions?** User access is managed through your existing Sammy Labs organization settings. Slack workspace permissions determine who can install/uninstall the bot. **Is the bot always active in our workspace?** The bot only activates when explicitly invoked via @mention. It performs no background processing or monitoring. **What if our team uses multiple Slack workspaces?** Each Slack workspace requires separate installation, but all integrate with your single Sammy Labs organization account. **How are videos processed?** Videos are downloaded, uploaded to secure S3 storage, analyzed by Vertex AI, and immediately deleted from temporary storage after processing. *** ## Getting Started Ready to transform your Slack workspace into an intelligent knowledge and policy hub? Connect your Slack workspace to Sammy Labs in just a few clicks. Requires Slack admin permissions and an active organization account. After installation, add the bot to your desired channels using `/invite @SAMMY` and configure your notification channel to start receiving policy updates.