⚙️Settings

Configure system settings, manage credentials, and review security audit logs

Configure your Automatos AI Platform - system, credentials, and security

---

📖 Table of Contents

1. Overview

2. Quick Start

3. System Settings Tab

4. Credentials Tab

5. Credential Types Tab

6. Audit Logs Tab

7. Common Tasks

8. Tips & Best Practices

9. Troubleshooting

---

Overview

What is Settings?

The Settings page is your control panel for platform configuration, credential management, and security auditing.

Access: Navigate to Settings from the sidebar (gear icon)

Settings Overview

What Can You Do Here?

• ✅ Configure system settings (6 sub-sections)

• ✅ Manage credentials securely (encrypted storage)

• ✅ Browse 400+ credential types

• ✅ Review audit logs for compliance

• ✅ Set API keys for LLM providers

• ✅ Configure logging and rate limiting

Page Layout

Settings has 4 main tabs:

1. ⚙️ System Settings - Platform configuration (6 sub-tabs)

2. 🔐 Credentials - Secure credential management

3. 📋 Credential Types - 400+ credential type templates

4. 📜 Audit Logs - Security and compliance logging

---

Quick Start

Configure Your First Credential (5 Minutes)

Goal: Add OpenAI API key for agents

Steps:

1. Go to Settings → Credentials tab

2. Click "Create Credential"

3. Type: Select "OpenAI API"

4. Name: "Production OpenAI"

5. API Key: Paste your OpenAI key

6. Test Connection: Verify it works ✅

7. Click "Save"

⏱️ Time: 3 minutes 🎯 Result: Agents can use OpenAI

See detailed walkthrough →

---

System Settings Tab

Overview

The System Settings tab has 6 sub-tabs for different configuration areas:

System Settings Tab

Sub-Tab 1: General

💡 Tooltip: "Basic platform configuration and preferences"

Platform Information (read-only):

• Platform Version: Current version

• API Version: API version number

• Installation ID: Unique instance ID

• License Type: Open Source / Enterprise

General Settings:

Platform Name 💡 Tooltip: "Custom name for your instance. Shown in UI and emails."

• Default: "Automatos AI Platform"

• Customize: "Acme Corp AI Platform"

Time Zone 💡 Tooltip: "Default timezone for timestamps and scheduling"

• Select from list

• Example: "America/New_York"

• Affects logs, schedules

Language 💡 Tooltip: "UI language. More languages coming soon."

• English (default)

• Future: Spanish, French, German, etc.

Theme 💡 Tooltip: "UI color scheme"

• Light mode

• Dark mode (default)

• Auto (system preference)

General Settings

Default Execution Settings:

Default Timeout 💡 Tooltip: "Default timeout for agent tasks and workflows. Can override per agent/workflow."

• Range: 30s to 30 minutes

• Recommended: 5 minutes

Default Token Budget 💡 Tooltip: "Default token limit for agent responses"

• Range: 100-8000 tokens

• Recommended: 2000 tokens

Auto-Start Agents 💡 Tooltip: "Automatically activate agents on platform startup"

• ✅ Enabled: Agents start automatically

• ❌ Disabled: Manual activation required

Sub-Tab 2: Orchestrator LLM

💡 Tooltip: "Configure the LLM used by the orchestrator for task decomposition and agent selection"

Orchestrator LLM Settings

Orchestrator Model:

Provider 💡 Tooltip: "LLM provider for orchestrator reasoning. Separate from agent models."

• OpenAI (default)

• Anthropic

• HuggingFace

Model 💡 Tooltip: "Specific model for orchestration. Needs function calling support."

• GPT-4 Turbo (recommended)

• GPT-4

• Claude 3 Opus

Temperature 💡 Tooltip: "Orchestrator creativity. Lower = more consistent decomposition."

• Range: 0.0-1.0

• Recommended: 0.3 (consistent)

Max Tokens 💡 Tooltip: "Max tokens for orchestrator responses (task decomposition, agent selection reasoning)"

• Range: 1000-4000

• Recommended: 2000

Reasoning Mode:

LLM-Driven Selection 💡 Tooltip: "Orchestrator uses reasoning to select agents. Higher quality but slower."

• ✅ Enabled (recommended)

• ❌ Disabled (falls back to algorithm)

Algorithmic Fallback 💡 Tooltip: "Use algorithm if LLM selection fails"

• ✅ Enabled for reliability

Sub-Tab 3: CodeGraph

💡 Tooltip: "Configure code indexing and search capabilities"

CodeGraph Settings

Indexing Settings:

Enabled 💡 Tooltip: "Enable/disable CodeGraph feature platform-wide"

• ✅ Enabled

• ❌ Disabled

Supported Languages 💡 Tooltip: "Programming languages to index"

• Python ✅

• TypeScript ✅

• JavaScript ✅

• Java ✅

• Go ✅

• Rust ✅

• C++ ✅

• Add custom languages

Max File Size 💡 Tooltip: "Maximum file size to index (larger files skipped)"

• Range: 100KB - 10MB

• Recommended: 1MB

Exclude Patterns 💡 Tooltip: "Glob patterns to skip during indexing"

• Default: node_modules, __pycache__, venv, .git, *.pyc

• Add more as needed

Cache Settings:

Cache TTL 💡 Tooltip: "How long to cache search results"

• Range: 0-3600 seconds

• Recommended: 300s (5 minutes)

Auto-Reindex 💡 Tooltip: "Automatically re-index on code changes (via webhooks)"

• ✅ Enabled for active projects

• ❌ Disabled for archived projects

Sub-Tab 4: Logging

💡 Tooltip: "Configure system logging and retention"

Logging Settings

Log Levels:

Application Log Level 💡 Tooltip: "Minimum severity to log. DEBUG = everything, ERROR = only errors."

• DEBUG (development)

• INFO (default)

• WARNING (production)

• ERROR (minimal)

Agent Log Level 💡 Tooltip: "Logging for agent executions"

• Same options as application

Workflow Log Level 💡 Tooltip: "Logging for workflow orchestration"

• Same options

Log Retention:

Keep Logs For 💡 Tooltip: "How long to retain logs before auto-deletion"

• 7 days

• 30 days (default)

• 90 days

• 1 year

• Forever

Log Storage Limit 💡 Tooltip: "Maximum disk space for logs. Oldest deleted when limit reached."

• Range: 1GB - 100GB

• Recommended: 10GB

Log Destinations:

• ✅ Database (searchable)

• ✅ File system (for export)

• ❌ External service (Datadog, etc.) - configure separately

Sub-Tab 5: Rate Limiting

💡 Tooltip: "Protect against API abuse and control costs"

Rate Limiting Settings

Global Rate Limits:

API Requests per Minute 💡 Tooltip: "Maximum API calls per minute (all users combined)"

• Range: 100-10,000

• Default: 1,000

• Protects against overload

Workflow Executions per Hour 💡 Tooltip: "Maximum workflows that can be created per hour"

• Range: 10-1,000

• Default: 100

• Prevents runaway automation

Agent Executions per Minute 💡 Tooltip: "Maximum agent tasks per minute"

• Range: 10-500

• Default: 50

Per-User Limits:

User API Rate

• Requests per minute per user

• Default: 100

User Workflow Rate

• Workflows per hour per user

• Default: 20

LLM Provider Limits:

OpenAI Requests per Minute 💡 Tooltip: "Limit to stay within OpenAI tier limits and control costs"

• Based on your OpenAI tier

• Default: 3,500 (Tier 3)

Anthropic Requests per Minute

• Based on Anthropic limits

• Default: 1,000

Sub-Tab 6: API Keys

💡 Tooltip: "Manage API keys for external access to Automatos AI"

API Keys Settings

Your API Keys:

List of active API keys:

• Key Name: Identifier

• Key Preview: First/last 8 characters (middle hidden)

• Permissions: Scopes assigned

• Last Used: Timestamp

• Created: Creation date

• Expires: Expiration date (if set)

Create API Key:

Click "Create API Key":

Name (required) 💡 Tooltip: "Descriptive name for this key. Example: 'CI/CD Pipeline', 'Mobile App'"

Permissions (checkboxes) 💡 Tooltip: "Grant specific permissions. Use least privilege principle."

• ✅ Read Agents

• ✅ Execute Agents

• ✅ Create Workflows

• ✅ Read Documents

• ❌ Delete Resources (rarely needed)

• ❌ Admin (very restricted)

Expiration 💡 Tooltip: "Auto-expire for security. Recommended: 90 days."

• Never (not recommended)

• 30 days

• 90 days (recommended)

• 1 year

• Custom date

Rate Limit 💡 Tooltip: "Override default rate limit for this key"

• Use default

• Custom: X requests/minute

Created Key:

• Shows full key ONCE

• Copy immediately

• Cannot view again (security)

• Store securely

Revoking Keys:

• Click "Revoke" button

• Immediate effect

• Cannot be undone

• All requests using key will fail

---

Credentials Tab

Overview

Securely manage authentication credentials for external services.

💡 Tooltip: "Encrypted storage for API keys, passwords, tokens. Required for agents to use external tools."

Credentials Tab

Statistics Cards

🔐 Total Credentials 💡 Tooltip: "Number of stored credentials. All encrypted at rest."

• Count of credentials

• Active vs inactive

✅ Verified 💡 Tooltip: "Credentials tested and working"

• Count with successful tests

• Percentage verified

⚠️ Expiring Soon 💡 Tooltip: "Credentials nearing expiration (if configured)"

• Count expiring <30 days

• Attention needed

🔧 In Use 💡 Tooltip: "Credentials actively assigned to agents"

• Count assigned to agents

• Usage percentage

Credentials Stats

Credential List

Each credential shows:

Credential Card:

• Name: Descriptive name (e.g., "Production OpenAI")

• Type: Credential type (OpenAI API, PostgreSQL, etc.)

• Status Badge:

  • 🟢 Active & Verified

  • 🟡 Active & Unverified

  • 🔴 Inactive or Failed Test

• Environment: Production / Development / Staging

• Created: Date added

• Last Used: Recent usage timestamp

Metadata:

• Tags (for organization)

• Description

• Assigned to: X agents

Credential Card

Actions:

• 🔧 Edit: Modify credential

• 🧪 Test: Test connection

• 👁️ View: See details (values hidden)

• 🗑️ Delete: Remove credential (with confirmation)

Creating a Credential

Click "Create Credential" button:

Create Credential Modal:

Create Credential Modal

Step 1: Select Type 💡 Tooltip: "Choose credential type. Determines required fields."

• Search or browse credential types

• Popular: OpenAI API, PostgreSQL, Slack, GitHub, AWS

• Shows required fields preview

• Select type

Step 2: Configure:

Credential Name (required) 💡 Tooltip: "Descriptive name. Example: 'Production OpenAI', 'Dev Database'"

Environment 💡 Tooltip: "Separate credentials by environment for safety"

• Production

• Development

• Staging

• Testing

Description (optional) 💡 Tooltip: "Notes about this credential - when to use, special configuration, etc."

Tags (optional):

• Add tags for organization

• Example: "ai", "production", "primary"

Credential Fields (depends on type):

Example: OpenAI API:

• API Key (required, secure) 🔒 💡 Tooltip: "OpenAI API key starting with 'sk-'. Get from platform.openai.com"

• Organization ID (optional)

• Base URL (optional, default: https://api.openai.com/v1)

Example: PostgreSQL:

• Host (required)

• Port (default: 5432)

• Database (required)

• User (required)

• Password (required, secure) 🔒

• SSL Mode (prefer, require, disable)

Example: Slack:

• Bot Token (required, secure) 🔒 💡 Tooltip: "Slack Bot User OAuth Token (xoxb-...)"

• Signing Secret (optional)

• App ID (optional)

Credential Fields

Step 3: Test Connection:

💡 Tooltip: "Verify credential works before saving. Highly recommended!"

• Click "Test Connection" button

• System attempts connection

• ✅ Success: Ready to save

• ❌ Failure: Shows error, fix before saving

Step 4: Save:

• Click "Save Credential"

• Encrypted and stored

• Available for tool assignments

Editing Credentials

Security Note: Existing values are hidden (shown as •••••)

When Editing:

• All fields shown but values hidden

• Update only fields you want to change

• Leave others blank to keep existing

• Re-test after editing

Testing Credentials

Click "Test" button on credential card:

Test Connection Process:

1. Decrypts credential

2. Attempts connection to service

3. Runs validation query/request

4. Returns result

Test Results:

✅ Success:

✅ Connection successful
Response time: 234ms
Test query: OK
Status: Verified

❌ Failure:

❌ Connection failed
Error: Invalid API key
Details: 401 Unauthorized
Action: Verify API key is correct

---

Credential Types Tab

Overview

Browse 400+ credential type templates.

💡 Tooltip: "Pre-configured credential schemas for every major service. Don't create from scratch!"

Credential Types Tab

Statistics

📋 Total Types: 417 credential types 📂 Categories: 12 categories ⭐ Most Used: OpenAI API (34 instances)

Browsing Credential Types

Category Filter:

💡 Tooltip: "Filter by service category"

• All Categories (417 types)

• AI & ML (12 types)

  • OpenAI API

  • Anthropic API

  • HuggingFace API

  • Cohere API

  • etc.

• Databases (25 types)

  • PostgreSQL

  • MySQL

  • MongoDB

  • Redis

  • etc.

• Cloud Providers (38 types)

  • AWS

  • Azure

  • Google Cloud

  • DigitalOcean

  • etc.

• Communication (45 types)

  • Slack

  • Discord

  • Telegram

  • Email (various providers)

  • etc.

• Developer Tools (67 types)

  • GitHub

  • GitLab

  • Jira

  • Jenkins

  • etc.

Credential Type Categories

Search:

• Search by name: "Slack"

• Search by provider: "Google"

• Search by capability: "notification"

Credential Type Details

Click any type to see details:

Type Information:

• Name: Official type name

• Provider: Service provider

• Icon: Visual identifier

• Category: Organization category

• Description: What it's for

Required Fields:

• Field name

• Field type (string, password, number, etc.)

• Required vs optional

• Default values

• Validation rules

Example Fields (OpenAI API):

✅ api_key (string, secure, required)
   Pattern: ^sk-[a-zA-Z0-9]{48}$
   
🔵 organization_id (string, optional)
   Pattern: ^org-[a-zA-Z0-9]+$
   
🔵 base_url (string, optional)
   Default: https://api.openai.com/v1

Documentation Links:

• Provider documentation

• How to get credentials

• Setup instructions

Usage Count:

• How many credentials of this type exist

• "34 credentials using this type"

---

Audit Logs Tab

Overview

Complete security audit trail of all credential operations.

💡 Tooltip: "Every credential access logged for compliance and security. SOC2, GDPR compliant."

Audit Logs Tab

Log Filters

Time Range 💡 Tooltip: "Filter by date range"

• Last 24 hours

• Last 7 days

• Last 30 days

• Last 90 days

• Custom range

Action Type 💡 Tooltip: "Filter by operation type"

• All Actions

• Created

• Updated

• Deleted

• Accessed (read)

• Tested

User Filter 💡 Tooltip: "Filter by who performed the action"

• All Users

• Select specific user

Credential Filter 💡 Tooltip: "Filter by which credential"

• All Credentials

• Select specific credential

Result Filter:

• All results

• Success only

• Failure only

Audit Log Filters

Audit Log Entries

Each log entry shows:

Log Record:

• Timestamp: Exact date/time

• Action: What happened

  • Created credential

  • Updated credential

  • Deleted credential

  • Accessed credential

  • Tested connection

• User: Who performed action

• Credential: Which credential (name + type)

• Result: ✅ Success or ❌ Failure

• IP Address: Source IP

• Details: Additional context

Audit Log Entry

Example Entries:

🟢 2024-10-21 10:35:28 | SUCCESS
   Action: Created credential
   User: [email protected]
   Credential: Production OpenAI (OpenAI API)
   IP: 192.168.1.100
   Details: New credential created via UI

🟡 2024-10-21 10:36:12 | SUCCESS
   Action: Tested connection
   User: [email protected]
   Credential: Production OpenAI
   IP: 192.168.1.100
   Details: Connection test successful (245ms)

🟢 2024-10-21 10:37:45 | SUCCESS
   Action: Accessed credential
   User: system
   Credential: Production OpenAI
   Service: agent-executor
   Details: Retrieved for agent CodeReviewer-Pro (agent_id: 42)

🔴 2024-10-21 11:15:33 | FAILURE
   Action: Tested connection
   User: [email protected]
   Credential: Slack Production
   IP: 192.168.1.100
   Details: Connection failed - Invalid token

Audit Log Export

Export Options:

💡 Tooltip: "Export audit logs for compliance reporting or analysis"

• CSV Export: Excel-compatible

• JSON Export: Programmatic analysis

• PDF Report: Human-readable audit report

Export Configuration:

• Date range

• Include/exclude fields

• Filter by criteria

• Format and download

---

Common Tasks

Task 1: Adding OpenAI API Key

Scenario: Configure OpenAI for agents

Steps:

1. Get API Key:

   • Go to platform.openai.com

   • Create API key

   • Copy key (starts with sk-)

2. Add to Automatos:

   • Settings → Credentials

   • Click "Create Credential"

   • Type: "OpenAI API"

   • Name: "Production OpenAI"

   • Paste API key

   • Test connection ✅

   • Save

3. Verify:

   • Agents can now use OpenAI models

   • Create test agent

   • Execute sample task

⏱️ Time: 5 minutes 🎯 Result: OpenAI configured

Task 2: Setting Up Slack Notifications

Scenario: Agents can send Slack messages

Steps:

1. Create Slack App:

   • Go to api.slack.com/apps

   • Create new app

   • Add Bot Token Scopes: chat:write, channels:read

   • Install app to workspace

   • Copy Bot User OAuth Token (xoxb-...)

2. Add to Automatos:

   • Settings → Credentials

   • Create "Slack" credential

   • Name: "Slack Production"

   • Bot Token: Paste token

   • Test ✅

   • Save

3. Assign to Agent:

   • Tools page → Slack

   • Assign to agent

   • Select "Slack Production" credential

   • Permissions: ✅ Write

   • Save

4. Test:

   • Execute agent task with instruction to send Slack message

   • Check Slack for message

⏱️ Time: 10 minutes 🎯 Result: Slack notifications working

Task 3: Configuring Database Credentials

Scenario: Agents need database access

Steps:

1. Settings → Credentials

2. Create "PostgreSQL" credential

3. Configure:

   • Name: "App Database"

   • Host: db.company.com

   • Port: 5432

   • Database: production_db

   • User: app_user

   • Password: •••••••• (secure)

   • SSL Mode: require

4. Test connection ✅

5. Save

6. Assign PostgreSQL tool to agents that need it

⏱️ Time: 5 minutes 🎯 Result: Database access configured

Task 4: Reviewing Audit Logs

Scenario: Monthly compliance review

Steps:

1. Settings → Audit Logs

2. Time Range: Last 30 days

3. Review:

   • All credential creations

   • All deletions

   • Failed access attempts

   • Unusual activity

4. Export:

   • Click "Export"

   • Format: PDF Report

   • Download for records

⏱️ Time: 10 minutes 🎯 Result: Compliance documentation

Task 5: Rotating Credentials

Scenario: Regular 90-day credential rotation

Steps:

1. Create new credential:

   • Generate new API key at provider

   • Create new credential in Automatos

   • Name: "OpenAI Production v2"

   • Test ✅

2. Update agent assignments:

   • Tools page

   • For each agent using old credential

   • Edit assignment

   • Change to new credential

   • Save

3. Verify:

   • Test agent execution

   • Ensure using new credential

4. Delete old credential:

   • Settings → Credentials

   • Find old credential

   • Verify not in use (0 assignments)

   • Delete

⏱️ Time: 15 minutes for 5 agents 🎯 Result: Credentials rotated securely

---

Advanced Features

Credential Templates

🔧 Advanced

Create custom credential types for internal services:

Custom Credential Type:

{
  "name": "Internal API",
  "provider": "Company Name",
  "category": "custom",
  "fields": [
    {
      "name": "api_endpoint",
      "type": "string",
      "required": true,
      "default": "https://internal.company.com/api"
    },
    {
      "name": "api_key",
      "type": "password",
      "required": true,
      "secure": true
    },
    {
      "name": "client_id",
      "type": "string",
      "required": false
    }
  ]
}

Environment Separation

Best Practice: Separate credentials per environment

Setup:

Production credentials:

• "Production OpenAI"

• "Production Database"

• "Production Slack"

Development credentials:

• "Development OpenAI"

• "Development Database"

• "Development Slack"

Benefits:

• No accidental prod access from dev

• Separate billing

• Different rate limits

• Clear audit trail

Credential Sharing

🔧 Advanced (Enterprise feature)

Share credentials across team:

Team Credential:

• Marked as "Shared"

• All team members can view/use

• Only admins can edit/delete

• All access logged per user

Personal Credential:

• Private to creator

• Others can't view/use

• Full control

---

Tips & Best Practices

Credential Naming

💡 Best Practices:

Good names:

• ✅ "Production OpenAI - Primary"

• ✅ "Dev Database - PostgreSQL"

• ✅ "Slack - Engineering Channel"

Poor names:

• ❌ "OpenAI 1"

• ❌ "Test"

• ❌ "My API Key"

Credential Security

Security Checklist:

1. ✅ Use Settings → Credentials (not .env files)

2. ✅ Test before using (verify they work)

3. ✅ Set expiration (for API keys)

4. ✅ Rotate regularly (90 days recommended)

5. ✅ Delete unused (reduce attack surface)

6. ✅ Monitor audit logs (catch unauthorized access)

7. ✅ Backup encryption key (.credential_key file)

Testing Credentials

Always test:

• ✅ After creation

• ✅ After editing

• ✅ Before production use

• ✅ After API key rotation

• ✅ If tool assignment fails

💡 Tip: Test connection button is your friend!

---

Troubleshooting

Credential Test Failed

Symptom: "Connection test failed" with error

Common Errors & Solutions:

"Invalid API key":

• Copy error (re-copy key)

• Expired key (regenerate)

• Wrong key type (check docs)

"Authentication failed":

• Wrong username/password

• Account locked/expired

• 2FA required (not supported)

"Connection timeout":

• Network issue

• Firewall blocking

• Wrong host/port

"Permission denied":

• Insufficient scopes

• Wrong access level

• Need admin approval

"Rate limit exceeded":

• Too many test attempts

• Wait a few minutes

• Check provider limits

Credential Not Appearing in Dropdown

Symptom: Created credential but can't select it

Solutions:

1. Check credential type:

   • Must match tool type

   • GitHub tool needs GitHub credential

   • Can't use Slack credential for GitHub

2. Check status:

   • Must be active

   • Verify not deleted

3. Refresh page:

   • Browser cache issue

   • Hard refresh (Ctrl/Cmd + Shift + R)

Can't Delete Credential

Symptom: "Cannot delete - credential in use"

Solutions:

1. Find assignments:

   • Credential details show agents using it

   • Go to each agent

   • Remove or replace credential

2. After removal:

   • Verify 0 assignments

   • Delete credential

⚠️ Safety: System prevents deleting credentials that would break agent tools.

Audit Logs Not Showing

Symptom: Audit tab is empty

Solutions:

1. Check time range:

   • May be filtering to period with no activity

   • Try "Last 90 days"

2. Check filters:

   • All filters set to "All"?

   • Clear filters

3. Verify logging enabled:

   • System Settings → Logging

   • Audit logging enabled?

---

Related Guides

• Tools Guide: Use credentials with tools

• Agents Guide: Tools require credentials

• Dashboard Guide: System health overview

---

Keyboard Shortcuts

• Ctrl/Cmd + K: Search credentials

• Ctrl/Cmd + N: New credential

• Ctrl/Cmd + T: Test selected credential

• Tab: Navigate between tabs

• Esc: Close modals

---

FAQ

Where are credentials stored?

Database (encrypted):

• PostgreSQL credentials table

• encrypted_data column (Fernet encryption)

• Encryption key in .credential_key file

• ⚠️ BACKUP .credential_key - losing it = losing all credentials

Can I import credentials from .env?

Yes! Migration script available:

cd orchestrator
python scripts/seed_credentials_from_env.py --dry-run
python scripts/seed_credentials_from_env.py

See Credential System Guide →

How do I backup credentials?

Backup encryption key:

cp orchestrator/.credential_key ~/secure_backup/
chmod 600 ~/secure_backup/.credential_key

Store in:

• Password manager

• Encrypted backup system

• Separate secure server

• ❌ NOT in git repository

Can multiple agents use same credential?

Yes! Common pattern:

• Single "Production OpenAI" credential

• Used by 10+ agents

• All share rate limits

• All usage tracked per agent

OR separate credentials:

• "CodeReviewer OpenAI"

• "SecurityAuditor OpenAI"

• Separate billing

• Separate rate limits

What happens if credential expires?

Auto-detection (some providers):

• System tests credentials periodically

• Failed test → status changes to "Failed"

• Notification sent

• Agents using it will fail with clear error

Manual monitoring:

• Check audit logs for failures

• Regular testing recommended

• Set expiration dates proactively

---

Next: Analytics Guide →

Monitor and optimize platform performance

---

API Reference

Tenants

Create, rename, delete tenants to isolate data & configs.

API

Authentication All API calls require headers:

X-API-Key: <your_key>
Authorization: Bearer <your_token>

• GET /api/settings/tenants

• POST /api/settings/tenants

• PUT /api/settings/tenants/{id}

• DELETE /api/settings/tenants/{id}

Providers

Store provider keys; set default models.

API

• GET /api/settings/providers

• PUT /api/settings/providers

Feature Flags

Toggle modules (Playbooks, Code Graph, A/B).

API

• GET /api/settings/flags

• PUT /api/settings/flags/{key} (body: {"value":true})

CORS

Manage allowed origins.

API

• GET /api/settings/cors

• PUT /api/settings/cors (body: {"origins":["http://localhost:3000"]})

Audit

Export audit logs.

API

• GET /api/settings/audit/info

• POST /api/settings/audit/export → {"url":"https://.../audit-2025-08-10.ndjson"}