Skip to Content
DojOps: AI-powered DevOps automation. Learn more →
Introduction

DojOps Documentation

Welcome to the DojOps documentation. DojOps generates, validates, and applies infrastructure and CI/CD configurations using LLM providers. Every output is schema-validated, sandboxed, and logged to a hash-chained audit trail before it touches your repo.



Documentation Index

Core Guides

DocumentDescription
Getting StartedPrerequisites, installation, provider setup, first run
Provider ManagementAdd, remove, switch, and list LLM providers
ConfigurationProviders, models, env vars, profiles, precedence
CLI ReferenceAll commands, flags, exit codes, examples
API ReferenceREST endpoints, request/response schemas, curl examples
Web Dashboard5-tab metrics dashboard, Overview, Security, Audit, Agents, History

Tutorials

DocumentDescription
Enhance a Next.js ProjectScan, analyze, and improve an existing project’s DevOps setup
CI/CD from ScratchBuild a full CI/CD pipeline from zero using plan + apply
Security Audit & RemediationSecurity scanning, SBOM generation, and auto-fix workflow
Interactive Chat SessionsMulti-turn AI conversations with slash commands and agent pinning
CI Debugging & AnalysisDebug CI failures, analyze diffs, auto-fix quality issues
Custom Agents & SkillsCreate agents, build skills, publish to DojOps Hub
Advanced Plan ExecutionDry run, resume, retry, replay, rollback, and targeted execution
API Server & DashboardSelf-host the API server, dashboard, TLS, and HTTP integration
Provider & ConfigurationManage providers, profiles, toolchain, and system health
Model Aliases & ThinkingCreate model aliases and control reasoning effort levels
Token Analytics & CostMonitor token usage, optimize costs, compression strategies
Advanced ConfigConfig backup/restore, import/export, validation, vault
Observability & LearningInsights, error patterns, memory, auto-enrichment
Autonomous AgentReAct loop, tool-call traces, iteration control, background runs
MCP External ToolsExtend the agent with external tools via Model Context Protocol
Voice InputSet up voice input and speak commands instead of typing

Architecture & Design

DocumentDescription
ArchitectureSystem design, package layers, dependency flow, data flow
Security ModelDefense-in-depth layers, trust boundary, concurrency safety

Components

DocumentDescription
Specialist Agents32 built-in + custom agents, routing, domains, keywords, confidence scoring, custom agent creation
DevOps SkillsAll 38 built-in skills + custom skill system, detectors, verifiers, auto-detection, skill format
Security Scanning10 scanners, scan types, remediation, SBOM, license compliance, scan comparison, dashboard integration
Execution EngineSafeExecutor, policies, approval workflows, sandboxed fs, audit trail
Task PlannerGoal decomposition, task graphs, topological execution, resume
MCPModel Context Protocol integration, transports, tool discovery, config

Operations

DocumentDescription
ContributingDev setup, coding standards, testing, adding skills/agents, PR workflow
TroubleshootingCommon issues, FAQ, error codes, debugging tips

Privacy & Telemetry

DojOps does not collect telemetry. No project data leaves your machine except to your configured LLM provider. All generated configs, audit logs, and scan reports are stored locally in your .dojops/ directory.