🚀 Complete Sloth Runner Features¶

Overview¶

Complete documentation of all Sloth Runner features - from basic capabilities to advanced enterprise functionality. This guide serves as the master index for exploring all platform capabilities.

📋 Feature Index¶

🎯 Core¶

Workflow Execution¶

Description: Central engine for executing workflows defined in Sloth files.

Characteristics: - Sequential and parallel task execution - Task group support - Variables and templating - Conditional execution - Error handling and retry - Dry-run mode - Verbose output

Commands:

sloth-runner run <workflow> --file <file>
sloth-runner run <workflow> --file <file> --yes
sloth-runner run <workflow> --file <file> --group <group>
sloth-runner run <workflow> --file <file> --values vars.yaml

Examples:

# Basic workflow
tasks:
  - name: Install nginx
    exec:
      script: |
        pkg.update()
        pkg.install("nginx")

  - name: Configure nginx
    exec:
      script: |
        file.copy("/src/nginx.conf", "/etc/nginx/nginx.conf")
        systemd.service_restart("nginx")

Documentation: /docs/en/quick-start.md

Sloth DSL Language¶

Description: Declarative DSL based on YAML with embedded Lua scripting.

Features: - YAML-based - familiar and readable syntax - Lua scripting - power of a complete language - Type-safe - type validation - Templating - Go templates and Jinja2 - Global modules - no require() needed - Modern syntax - supports modern features

Structure:

# Metadata
version: "1.0"
description: "My workflow"

# Variables
vars:
  env: production
  version: "1.2.3"

# Groups
groups:
  deploy:
    - install_deps
    - build_app
    - deploy_app

# Tasks
tasks:
  - name: install_deps
    exec:
      script: |
        pkg.install({"nodejs", "npm"})

  - name: build_app
    exec:
      script: |
        exec.command("npm install")
        exec.command("npm run build")

  - name: deploy_app
    exec:
      script: |
        file.copy("./dist", "/var/www/app")
        systemd.service_restart("app")
    delegate_to: web-01

Documentation: /docs/modern-dsl/introduction.md

Module System¶

Description: 40+ integrated modules providing all automation needs.

Categories:

📦 System¶

pkg - Package management (apt, yum, brew, etc.)
user - User/group management
file - File operations
systemd - Service management
exec - Command execution

🐳 Containers¶

docker - Complete Docker (containers, images, networks)
incus - Incus/LXC containers and VMs
kubernetes - K8s deploy and management

☁️ Cloud¶

aws - AWS (EC2, S3, RDS, Lambda, etc.)
azure - Azure (VMs, Storage, etc.)
gcp - GCP (Compute Engine, Cloud Storage, etc.)
digitalocean - DigitalOcean (Droplets, Load Balancers)

🏗️ IaC¶

terraform - Terraform (init, plan, apply, destroy)
pulumi - Pulumi
ansible - Ansible playbooks

🔧 Tools¶

git - Git operations
ssh - Remote SSH
net - Networking (ping, http, download)
template - Templates (Jinja2, Go)

📊 Observability¶

log - Structured logging
metrics - Metrics (Prometheus)
notifications - Notifications (Slack, Email, Discord, Telegram)

🚀 Advanced¶

goroutine - Parallel execution
reliability - Retry, circuit breaker, timeout
state - State management
facts - System information
infra_test - Infrastructure testing

Complete list: sloth-runner modules list

Documentation: /docs/en/modules-complete.md

State Management¶

Description: State persistence system between executions.

Features: - Persistent key-value store - SQLite backend - State scoping (global, workflow, task) - Change detection - State cleanup

API:

-- Save state
state.set("last_deploy_version", "v1.2.3")
state.set("deploy_timestamp", os.time())

-- Read state
local last_version = state.get("last_deploy_version")

-- Detect change
if state.changed("config_hash", new_hash) then
    log.info("Config changed, redeploying")
    deploy()
end

-- Clear state
state.clear("temporary_data")

Documentation: /docs/state-management.md

Idempotency¶

Description: Ensures workflows can be executed multiple times with the same result.

Features: - Check mode - checks before executing - State tracking - tracks what was changed - Resource fingerprinting - detects changes - Rollback - undoes changes on error

Example:

-- Idempotent - checks before installing
if not pkg.is_installed("nginx") then
    pkg.install("nginx")
end

-- Idempotent - checks file hash
local current_hash = file.hash("/etc/nginx/nginx.conf")
if current_hash ~= expected_hash then
    file.copy("/src/nginx.conf", "/etc/nginx/nginx.conf")
    systemd.service_restart("nginx")
end

Documentation: /docs/idempotency.md

🌐 Distributed¶

Master-Agent Architecture¶

Description: Distributed architecture with central master server and remote agents.

Components: - Master Server - coordinates agents and workflows - Agent Nodes - execute tasks remotely - gRPC Communication - efficient and type-safe communication - Auto-Discovery - agents self-register - Health Monitoring - automatic heartbeats

Topology:

                    ┌──────────────┐
                    │   Master     │
                    │  (gRPC:50053)│
                    └──────┬───────┘
                           │
         ┌─────────────────┼─────────────────┐
         │                 │                 │
    ┌────▼────┐       ┌────▼────┐      ┌────▼────┐
    │ Agent 1 │       │ Agent 2 │      │ Agent 3 │
    │  web-01 │       │  web-02 │      │   db-01 │
    └─────────┘       └─────────┘      └─────────┘

Setup:

# Start master
sloth-runner server --port 50053

# Install remote agent
sloth-runner agent install web-01 \
  --ssh-host 192.168.1.100 \
  --ssh-user ubuntu \
  --master 192.168.1.1:50053

# List agents
sloth-runner agent list

Documentation: /docs/en/master-agent-architecture.md

Task Delegation¶

Description: Delegates task execution to specific agents.

Features: - Single delegation - delegate to one agent - Multi delegation - delegate to multiple agents in parallel - Round-robin - distribute load - Failover - fallback if agent fails - Conditional delegation - delegate based on conditions

Syntax:

# Delegate to one agent
tasks:
  - name: Deploy to web-01
    exec:
      script: |
        pkg.install("nginx")
    delegate_to: web-01

# Delegate to multiple agents
tasks:
  - name: Deploy to all web servers
    exec:
      script: |
        pkg.install("nginx")
    delegate_to:
      - web-01
      - web-02
      - web-03

# CLI - delegate entire workflow
sloth-runner run deploy --file deploy.sloth --delegate-to web-01

Use with values:

# Pass agent-specific values
tasks:
  - name: Configure
    exec:
      script: |
        local ip = values.ip_address
        file.write("/etc/config", "IP=" .. ip)
    delegate_to: "{{ item }}"
    loop:
      - web-01
      - web-02
    values:
      web-01:
        ip_address: "192.168.1.10"
      web-02:
        ip_address: "192.168.1.11"

Documentation: /docs/guides/values-delegate-to.md

gRPC Communication¶

Description: Efficient communication between master and agents using gRPC.

Features: - Streaming - bi-directional streaming - Type-safe - Protocol Buffers - Efficient - binary protocol - Multiplexing - multiple calls in one connection - TLS - TLS/SSL support

Services:

service AgentService {
    rpc ExecuteTask(TaskRequest) returns (TaskResponse);
    rpc StreamLogs(LogRequest) returns (stream LogEntry);
    rpc Heartbeat(HeartbeatRequest) returns (HeartbeatResponse);
    rpc GetMetrics(MetricsRequest) returns (MetricsResponse);
}

Default port: 50053

Auto-Reconnection¶

Description: Agents automatically reconnect to master on disconnection.

Features: - Exponential backoff - increases interval between attempts - Max retries - configurable limit - Circuit breaker - stops trying after many failures - Connection pooling - reuses connections

Configuration:

agent:
  reconnect:
    enabled: true
    initial_delay: 1s
    max_delay: 60s
    max_retries: -1  # infinite

Documentation: /docs/en/agent-improvements.md

Health Checks¶

Description: Continuous agent health monitoring.

Check types: - Heartbeat - periodic ping - Resource check - CPU, memory, disk - Service check - checks critical services - Custom checks - user-defined checks

Endpoints:

# Health endpoint
curl http://agent:9090/health

# Metrics endpoint
curl http://agent:9090/metrics

Thresholds:

health:
  cpu_threshold: 90  # %
  memory_threshold: 85  # %
  disk_threshold: 90  # %
  heartbeat_interval: 30s
  heartbeat_timeout: 90s

🎨 Interface¶

Modern Web UI¶

Description: Complete, responsive, real-time web interface.

Main features: - ✅ Dashboard with metrics and charts - ✅ Agent management with real-time metrics - ✅ Workflow editor with syntax highlighting - ✅ Execution and log viewing - ✅ Interactive terminal (xterm.js) - ✅ Dark mode / Light mode - ✅ WebSocket for real-time updates - ✅ Mobile responsive - ✅ Command palette (Ctrl+Shift+P) - ✅ Drag & drop - ✅ Glassmorphism design - ✅ Smooth animations

Pages: 1. Dashboard (/) 2. Agents (/agents) 3. Agent Control (/agent-control) 4. Agent Dashboard (/agent-dashboard) 5. Workflows (/workflows) 6. Executions (/executions) 7. Hooks (/hooks) 8. Events (/events) 9. Scheduler (/scheduler) 10. Logs (/logs) 11. Terminal (/terminal) 12. Sloths (/sloths) 13. Settings (/settings)

Technologies: - Bootstrap 5.3 - Chart.js 4.4 - xterm.js - WebSockets - Canvas API

Start:

sloth-runner ui --port 8080

Access: http://localhost:8080

Documentation: /docs/en/web-ui-complete.md

Complete CLI¶

Description: Complete command-line interface with 100+ commands.

Command categories:

Execution¶

run - Execute workflow
version - View version

Agents¶

agent list - List agents
agent get - Agent details
agent install - Install remote agent
agent update - Update agent
agent start/stop/restart - Control agent
agent modules - List agent modules
agent metrics - View metrics

Sloths (Saved Workflows)¶

sloth list - List sloths
sloth add - Add sloth
sloth get - View sloth
sloth update - Update sloth
sloth remove - Remove sloth
sloth activate/deactivate - Activate/deactivate

Hooks¶

hook list - List hooks
hook add - Add hook
hook remove - Remove hook
hook enable/disable - Enable/disable
hook test - Test hook

Events¶

events list - List events
events watch - Monitor events in real-time

Database¶

db backup - Database backup
db restore - Restore database
db vacuum - Optimize database
db stats - Statistics

SSH¶

ssh list - List SSH connections
ssh add - Add connection
ssh remove - Remove connection
ssh test - Test connection

Modules¶

modules list - List modules
modules info - Module info

Server¶

server - Start master server
ui - Start Web UI
terminal - Interactive terminal

Utilities¶

completion - Shell auto-completion
doctor - Diagnostics

Documentation: /docs/en/cli-reference.md

Interactive REPL¶

Description: Read-Eval-Print Loop for interactive Lua code testing.

Features: - Complete Lua - full Lua interpreter - Loaded modules - all modules available - History - command history - Auto-complete - Tab completion - Multi-line - multi-line code support - Pretty print - formatted output

Start:

sloth-runner repl

Example session:

> pkg.install("nginx")
[OK] nginx installed successfully

> file.exists("/etc/nginx/nginx.conf")
true

> local content = file.read("/etc/nginx/nginx.conf")
> print(#content .. " bytes")
2048 bytes

> for i=1,5 do
>>   print("Hello " .. i)
>> end
Hello 1
Hello 2
Hello 3
Hello 4
Hello 5

Special commands: - .help - help - .exit - exit - .clear - clear screen - .load <file> - load file - .save <file> - save session

Documentation: /docs/en/repl.md

Remote Terminal¶

Description: Interactive terminal for remote agents via web UI.

Features: - xterm.js - complete terminal emulator - Multiple sessions - multiple simultaneous sessions - Tabs - tab management - Command history - command history (↑↓) - Copy/paste - Ctrl+Shift+C/V - Themes - various themes available - Upload/download - file transfer

Access: 1. Web UI → Terminal 2. Select agent 3. Connect

Special commands:

.clear       # Clear terminal
.exit        # Close session
.upload <f>  # Upload file
.download <f># Download file
.theme <t>   # Change theme

URL: http://localhost:8080/terminal

REST API¶

Description: Complete RESTful API for external integration.

Main endpoints:

Agents¶

GET    /api/v1/agents           # List agents
GET    /api/v1/agents/:name     # Agent details
POST   /api/v1/agents/:name/restart  # Restart agent
DELETE /api/v1/agents/:name     # Remove agent

Workflows¶

POST   /api/v1/workflows/run    # Execute workflow
GET    /api/v1/workflows/:id    # Workflow details

Executions¶

GET    /api/v1/executions       # List executions
GET    /api/v1/executions/:id   # Execution details

Hooks¶

GET    /api/v1/hooks            # List hooks
POST   /api/v1/hooks            # Create hook
DELETE /api/v1/hooks/:name      # Remove hook

Events¶

GET    /api/v1/events           # List events

Metrics¶

GET    /api/v1/metrics          # Prometheus metrics

Authentication:

curl -H "Authorization: Bearer <token>" \
  http://localhost:8080/api/v1/agents

Examples:

# List agents
curl http://localhost:8080/api/v1/agents

# Execute workflow
curl -X POST http://localhost:8080/api/v1/workflows/run \
  -H "Content-Type: application/json" \
  -d '{
    "file": "/workflows/deploy.sloth",
    "workflow_name": "deploy",
    "delegate_to": ["web-01"]
  }'

# View metrics
curl http://localhost:8080/api/v1/metrics

Documentation: /docs/web-ui/api-reference.md

🔧 Automation¶

Scheduler¶

Description: Cron-based workflow scheduler.

Features: - Cron expressions - complete cron syntax - Visual builder - visual builder in Web UI - Timezone support - timezone support - Missed run policy - policy for missed runs - Overlap prevention - prevents overlapping executions - Notifications - success/failure notifications

Create job:

# Via CLI (coming soon)
sloth-runner scheduler add deploy-job \
  --workflow deploy.sloth \
  --schedule "0 3 * * *"  # Every day at 3am

# Via Web UI
http://localhost:8080/scheduler

Cron syntax:

┌───────────── minute (0 - 59)
│ ┌───────────── hour (0 - 23)
│ │ ┌───────────── day of month (1 - 31)
│ │ │ ┌───────────── month (1 - 12)
│ │ │ │ ┌───────────── day of week (0 - 6) (Sunday=0)
│ │ │ │ │
* * * * *

Examples:
0 * * * *     # Every hour
0 3 * * *     # Every day at 3am
0 0 * * 0     # Every Sunday at midnight
*/15 * * * *  # Every 15 minutes

Documentation: /docs/scheduler.md

Hooks & Events¶

Description: Hook system for reacting to system events.

Available events: - workflow.started - Workflow started - workflow.completed - Workflow completed - workflow.failed - Workflow failed - task.started - Task started - task.completed - Task completed - task.failed - Task failed - agent.connected - Agent connected - agent.disconnected - Agent disconnected

Create hook:

sloth-runner hook add slack-notify \
  --event workflow.completed \
  --script /scripts/notify-slack.lua \
  --priority 10

Hook script (Lua):

-- /scripts/notify-slack.lua
local event = hook.event
local payload = hook.payload

if event == "workflow.completed" then
    notifications.slack(
        "https://hooks.slack.com/services/XXX/YYY/ZZZ",
        string.format("✅ Workflow '%s' completed!", payload.workflow_name),
        { channel = "#deployments" }
    )
end

Available payload:

-- workflow.* events
{
    workflow_name = "deploy",
    status = "success",
    duration = 45.3,
    started_at = 1234567890,
    completed_at = 1234567935
}

-- agent.* events
{
    agent_name = "web-01",
    address = "192.168.1.100:50060",
    status = "connected"
}

Documentation: /docs/architecture/hooks-events-system.md

GitOps¶

Description: Complete GitOps pattern implementation.

Features: - Git-based - Git as source of truth - Auto-sync - automatic synchronization - Drift detection - detects manual changes - Rollback - automatic rollback - Multi-environment - dev, staging, production - PR-based - approval via Pull Requests

GitOps workflow:

# .sloth/gitops.yaml
repos:
  - name: k8s-manifests
    url: https://github.com/org/k8s-manifests.git
    branch: main
    path: production/
    sync_interval: 5m
    auto_sync: true
    prune: true

hooks:
  on_sync:
    - notify-slack
  on_drift:
    - alert-team

CLI:

# Manual sync
sloth-runner gitops sync k8s-manifests

# View status
sloth-runner gitops status

# View drift
sloth-runner gitops diff

Documentation: /docs/en/gitops-features.md

CI/CD Integration¶

Description: Integration with CI/CD pipelines.

Support: - GitHub Actions - GitLab CI - Jenkins - CircleCI - Travis CI - Azure Pipelines

GitHub Actions example:

# .github/workflows/deploy.yml
name: Deploy

on:
  push:
    branches: [main]

jobs:
  deploy:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v2

      - name: Install Sloth Runner
        run: |
          curl -L https://github.com/org/sloth-runner/releases/latest/download/sloth-runner-linux-amd64 -o sloth-runner
          chmod +x sloth-runner

      - name: Run deployment
        env:
          SLOTH_RUNNER_MASTER_ADDR: ${{ secrets.SLOTH_MASTER }}
        run: |
          ./sloth-runner run deploy \
            --file deployments/production.sloth \
            --delegate-to web-01 \
            --yes

Sloths¶

Description: Repository of saved and reusable workflows.

Features: - Versioning - version history - Tags - organization by tags - Search - search by name/description/tags - Clone - clone existing sloth - Export/Import - share sloths - Active/Inactive - activate/deactivate without deleting

Commands:

# Add sloth
sloth-runner sloth add deploy --file deploy.sloth

# List sloths
sloth-runner sloth list

# View sloth
sloth-runner sloth get deploy

# Execute sloth
sloth-runner run deploy --file $(sloth-runner sloth get deploy --show-path)

# Remove sloth
sloth-runner sloth remove deploy

Documentation: /docs/features/sloth-management.md

📊 Monitoring¶

Telemetry¶

Description: Complete observability system.

Components: - Prometheus metrics - Structured logging - Distributed tracing - Health checks - Performance profiling

Architecture:

┌──────────┐    metrics    ┌────────────┐
│  Master  ├───────────────► Prometheus │
└──────────┘               └─────┬──────┘
                                 │
┌──────────┐    metrics          │
│ Agent 1  ├─────────────────────┤
└──────────┘                     │
                                 ▼
┌──────────┐    metrics    ┌──────────┐
│ Agent 2  ├───────────────►  Grafana │
└──────────┘               └──────────┘

Endpoints:

http://master:9090/metrics
http://agent:9091/metrics

Documentation: /docs/en/telemetry/index.md

Prometheus Metrics¶

Description: Metrics exported in Prometheus format.

Available metrics:

Workflows¶

sloth_workflow_executions_total{status="success|failed"}
sloth_workflow_duration_seconds{workflow="name"}
sloth_workflow_tasks_total{workflow="name"}

Agents¶

sloth_agent_connected_total
sloth_agent_cpu_usage_percent{agent="name"}
sloth_agent_memory_usage_bytes{agent="name"}
sloth_agent_disk_usage_bytes{agent="name"}

System¶

sloth_tasks_executed_total
sloth_hooks_triggered_total{event="type"}
sloth_db_size_bytes

Scrape config:

# prometheus.yml
scrape_configs:
  - job_name: 'sloth-master'
    static_configs:
      - targets: ['localhost:9090']

  - job_name: 'sloth-agents'
    static_configs:
      - targets:
        - 'agent1:9091'
        - 'agent2:9091'

Documentation: /docs/en/telemetry/prometheus-metrics.md

Grafana Dashboards¶

Description: Pre-configured dashboards for Grafana.

Dashboards: 1. Overview - system overview 2. Agents - metrics for all agents 3. Workflows - executions and performance 4. Resources - CPU, memory, disk, network

Import dashboard:

# Generate dashboard JSON
sloth-runner agent metrics grafana web-01 --export dashboard.json

# Import to Grafana
curl -X POST http://admin:admin@localhost:3000/api/dashboards/db \
  -H "Content-Type: application/json" \
  -d @dashboard.json

Features: - Auto-refresh (5s, 10s, 30s, 1m) - Time range selector - Variables (agent, workflow) - Configurable alerts - Export PNG/PDF

Documentation: /docs/en/telemetry/grafana-dashboard.md

Centralized Logs¶

Description: Centralized structured logging system.

Features: - Structured - JSON structured logs - Levels - debug, info, warn, error - Context - rich metadata - Search - search by any field - Export - JSON, CSV, text - Retention - retention policy

Format:

{
  "timestamp": "2025-10-07T10:30:45Z",
  "level": "info",
  "message": "Workflow completed",
  "workflow": "deploy",
  "agent": "web-01",
  "duration": 45.3,
  "status": "success"
}

Access:

# CLI
sloth-runner logs --follow

# Web UI
http://localhost:8080/logs

# API
curl http://localhost:8080/api/v1/logs?level=error&since=1h

Agent Metrics¶

Description: Detailed real-time agent metrics.

Collected metrics: - CPU usage (%) - Memory usage (bytes, %) - Disk usage (bytes, %) - Load average (1m, 5m, 15m) - Network I/O (bytes/sec) - Process count - Uptime

Visualization:

# CLI
sloth-runner agent metrics web-01
sloth-runner agent metrics web-01 --watch

# Web UI - Agent Dashboard
http://localhost:8080/agent-dashboard?agent=web-01

# API
curl http://localhost:8080/api/v1/agents/web-01/metrics

Format:

{
  "cpu": {
    "cores": 4,
    "usage_percent": 45.2,
    "load_avg": [1.2, 0.8, 0.5]
  },
  "memory": {
    "total_bytes": 8589934592,
    "used_bytes": 4294967296,
    "usage_percent": 50.0
  },
  "disk": {
    "total_bytes": 107374182400,
    "used_bytes": 53687091200,
    "usage_percent": 50.0
  }
}

☁️ Cloud & IaC¶

Multi-Cloud¶

Description: Native support for multiple cloud providers.

Supported providers: - ✅ AWS (EC2, S3, RDS, Lambda, ECS, EKS, etc.) - ✅ Azure (VMs, Storage, AKS, Functions, etc.) - ✅ GCP (Compute Engine, Cloud Storage, GKE, etc.) - ✅ DigitalOcean (Droplets, Spaces, K8s, etc.) - ✅ Linode - ✅ Vultr - ✅ Hetzner Cloud

Multi-cloud example:

# Deploy to AWS and GCP simultaneously
tasks:
  - name: Deploy to AWS
    exec:
      script: |
        aws.ec2_instance_create({
          image_id = "ami-xxx",
          instance_type = "t3.medium"
        })
    delegate_to: aws-agent

  - name: Deploy to GCP
    exec:
      script: |
        gcp.compute_instance_create({
          machine_type = "e2-medium",
          image_family = "ubuntu-2204-lts"
        })
    delegate_to: gcp-agent

Documentation: /docs/en/enterprise-features.md

Terraform¶

Description: Complete Terraform integration.

Features: - terraform.init - Initialize - terraform.plan - Plan - terraform.apply - Apply - terraform.destroy - Destroy - State management - Backend config - Variable files

Example:

local tf_dir = "/infra/terraform"

-- Initialize
terraform.init(tf_dir, {
    backend_config = {
        bucket = "my-tf-state",
        key = "prod/terraform.tfstate"
    }
})

-- Plan
local plan = terraform.plan(tf_dir, {
    var_file = "production.tfvars",
    vars = {
        region = "us-east-1",
        environment = "production"
    }
})

-- Apply if there are changes
if plan.changes > 0 then
    terraform.apply(tf_dir, {
        auto_approve = true
    })
end

Documentation: /docs/modules/terraform.md

Pulumi¶

Description: Pulumi integration.

Support: - Stack management - Configuration - Up/Deploy - Destroy - Preview

Example:

-- Select stack
pulumi.stack_select("production")

-- Configure
pulumi.config_set("aws:region", "us-east-1")

-- Deploy
pulumi.up({
    yes = true,  -- auto-approve
    parallel = 10
})

Documentation: /docs/modules/pulumi.md

Kubernetes¶

Description: Kubernetes deploy and management.

Features: - Apply manifests - Helm charts - Namespaces - ConfigMaps/Secrets - Rollouts - Health checks

Example:

-- Apply manifests
kubernetes.apply("/k8s/deployment.yaml", {
    namespace = "production"
})

-- Helm install
helm.install("myapp", "charts/myapp", {
    namespace = "production",
    values = {
        image = {
            tag = "v1.2.3"
        }
    }
})

-- Wait for rollout
kubernetes.rollout_status("deployment/myapp", {
    namespace = "production",
    timeout = "5m"
})

Documentation: /docs/en/gitops/kubernetes.md

Docker¶

Description: Complete Docker automation.

Functionality: - Container lifecycle (run, stop, remove) - Image management (build, push, pull) - Networks (create, connect) - Volumes (create, mount) - Docker Compose

Deployment example:

-- Build image
docker.image_build(".", {
    tag = "myapp:v1.2.3",
    build_args = {
        VERSION = "1.2.3"
    }
})

-- Push to registry
docker.image_push("myapp:v1.2.3", {
    registry = "registry.example.com"
})

-- Deploy
docker.container_run("myapp:v1.2.3", {
    name = "app",
    ports = {"3000:3000"},
    env = {
        DATABASE_URL = "postgres://..."
    },
    restart = "unless-stopped"
})

Documentation: /docs/modules/docker.md

🔐 Security & Enterprise¶

Authentication¶

Description: Authentication system for Web UI and API.

Methods: - Username/Password - JWT tokens - OAuth2 (GitHub, Google, etc.) - LDAP/AD - SSO

Setup:

# config.yaml
auth:
  enabled: true
  type: jwt
  jwt:
    secret: "your-secret-key"
    expiry: 24h
  oauth:
    providers:
      - github:
          client_id: "xxx"
          client_secret: "yyy"

TLS/SSL¶

Description: TLS/SSL support for secure communication.

Features: - gRPC TLS - HTTPS Web UI - Certificate management - Auto-renewal (Let's Encrypt)

Configuration:

# Master with TLS
sloth-runner server \
  --tls-cert /etc/sloth/cert.pem \
  --tls-key /etc/sloth/key.pem

# Agent with TLS
sloth-runner agent start \
  --master-tls-cert /etc/sloth/master-cert.pem

Audit Logs¶

Description: Audit logs for all actions.

Audited events: - User login/logout - Workflow execution - Configuration changes - API calls - Admin actions

Format:

{
  "timestamp": "2025-10-07T10:30:45Z",
  "event": "workflow.executed",
  "user": "admin",
  "ip": "192.168.1.100",
  "resource": "deploy.sloth",
  "action": "execute",
  "result": "success"
}

Backups¶

Description: Automated backup system.

Features: - Configurable auto-backup - Compression (gzip) - Retention policy - Remote backup (S3, Azure Blob, etc.) - Restore

Commands:

# Manual backup
sloth-runner db backup --output /backup/sloth.db --compress

# Restore
sloth-runner db restore /backup/sloth.db.gz --decompress

# Automated backup (cron)
0 3 * * * sloth-runner db backup --output /backup/sloth-$(date +\%Y\%m\%d).db --compress

RBAC¶

Description: Role-Based Access Control.

Roles: - Admin - full access - Operator - execute workflows, manage agents - Developer - create/edit workflows - Viewer - view only

Permissions:

id=__codelineno-54-1 name=__codelineno-54-1 href=#__codelineno-54-1>roles: operator: permissions: - workflow:execute - agent:view - agent:restart - logs:view developer: permissions: - workflow:create - workflow:edit - workflow:execute - logs:view viewer: permissions: - workflow:view - agent:view - logs:view

🚀 Performance¶

Optimizations¶

Description: Recent performance optimizations.

Implemented improvements:

Agent Optimizations¶

✅ Ultra-low memory - 32MB RAM footprint
✅ Binary size reduction - from 45MB → 12MB
✅ Startup time - <100ms
✅ CPU efficiency - 99% idle when inactive

Database Optimizations¶

✅ WAL mode - Write-Ahead Logging
✅ Connection pooling - connection reuse
✅ Prepared statements - optimized queries
✅ Indexes - indexes on critical fields
✅ Auto-vacuum - automatic cleanup

gRPC Optimizations¶

✅ Connection reuse - keepalive
✅ Compression - gzip compression
✅ Multiplexing - multiple streams
✅ Buffer pooling - buffer reuse

Benchmark:

Before:
- Agent memory: 128MB
- Binary size: 45MB
- Startup time: 2s

After:
- Agent memory: 32MB (75% reduction)
- Binary size: 12MB (73% reduction)
- Startup time: 95ms (95% faster)

Documentation: /PERFORMANCE_OPTIMIZATIONS.md

Parallel Execution¶

Description: Parallel task execution using goroutines.

Features: - goroutine.parallel() - execute functions in parallel - Concurrency control - limit of simultaneous goroutines - Error handling - collects errors from all goroutines - Wait groups - automatic synchronization

Example:

-- Execute multiple tasks in parallel
goroutine.parallel({
    function()
        pkg.install("nginx")
    end,
    function()
        pkg.install("postgresql")
    end,
    function()
        pkg.install("redis")
    end
})

-- With concurrency limit
goroutine.parallel({
    tasks = {
        function() exec.command("task1") end,
        function() exec.command("task2") end,
        function() exec.command("task3") end,
        function() exec.command("task4") end
    },
    max_concurrent = 2  -- Maximum 2 at a time
})

Documentation: /docs/modules/goroutine.md

Resource Limits¶

Description: Configurable resource limits.

Configuration:

# Agent config
resources:
  cpu:
    limit: 2  # cores
    reserve: 0.5
  memory:
    limit: 2GB
    reserve: 512MB
  disk:
    limit: 10GB
    min_free: 1GB

Enforcement: - CPU throttling - Memory limits (cgroup) - Disk quota - Task timeout

Caching¶

Description: Caching system for optimization.

Cache types:

Module cache¶

Compiled Lua modules
Reduce load time

State cache¶

State in memory
Reduce DB queries

Metrics cache¶

Aggregated metrics
Reduce computation

Configuration:

cache:
  enabled: true
  ttl: 5m
  max_size: 100MB
  eviction: lru  # least recently used

📚 Additional Resources¶

Documentation¶

Useful Links¶

Last updated: 2025-10-07

Total Documented Features: 100+