📄️ status
Show overall system status across all projects. Aggregates health, errors, and pipeline status from every registered project into a single view.
📄️ azure
On-demand queries of live Azure state — resource inventory, month-to-date costs, RBAC role assignments, and non-human identities. Every subcommand shells out to the authenticated az CLI and reflects current state; nothing is stored or cached.
📄️ pipeline
Pipeline management commands for listing, inspecting, triggering, and cancelling pipeline runs across the portfolio.
📄️ prefect
Prefect flow and deployment management. Prefect is the primary flow orchestrator for all data pipelines in the Wasteology portfolio. These commands are now the single CLI surface for the complete flow lifecycle — an agent (or engineer) can probe an API, smoke-test locally, ship a new image, deploy, run with row-count gating, and diagnose stalls entirely through ops prefect without touching the Prefect UI or bare az / prefect CLIs.
📄️ dbt
dbt data transformation operations. All commands run against the dbt Cloud CLI (not dbt-core). The dbt project lives at ~/projects/data-warehouses/dbtwasteologywarehouse.
📄️ sling
Sling database operations and queries. Sling is used to query SQL Server (CieTrade, Azure SQL) databases. Results stream directly to the terminal.
📄️ n8n
n8n workflow management. n8n is the automation platform used for webhook triggers, scheduled workflows, and integration glue between services.
📄️ digest
Daily digest generation, reporting, and task management. The digest system tracks project health over time and generates AI-powered summaries for standups and stakeholder communication.
📄️ portfolio
Portfolio management and cross-project visibility across all Wasteology projects. Provides unified status, git state, and strategic context queries across every registered project.
📄️ project
Project registry management. Projects are defined in TOML manifests at manifests/builtin/*.toml. This command group is used to inspect, validate, and diagnose individual projects.
📄️ db
PostgreSQL database user and role management. All commands connect to the wg_digests PostgreSQL database. Supports idempotent user creation, RBAC role management, and grant auditing.
📄️ agent
Multi-agent orchestration. Broadcasts tasks to project master agents (Claude Code instances) in parallel, chains agents sequentially, and manages saved chain configs. Each project's CLAUDE.md makes a spawned agent an instant expert on that project.
📄️ ai
AI assistant commands. Provides a unified interface to Claude (via the Claude Code CLI) and Google Gemini (via the google-genai API). Use auto to let the CLI pick the best available provider.
📄️ orch
Orchestrator Agent interface. Connects to the ADW (Autonomous DevOps Worker) orchestrator agent for triggering and monitoring ADW workflows. The ADW pipeline processes digest tasks through planning, implementation, and PR creation.
📄️ monitor
Continuous project monitoring commands. Polls project health and surfaces issues in real time.
📄️ docs
Documentation site management. Manages deployment and freshness of the Docusaurus documentation sites hosted on Azure Container Apps. There are two sites: internal (this site) and crm (vendor-facing CieTrade guide).
📄️ tech-stack
Generate tech stack visualization images for projects. Images are defined in .tech-stack.yaml config files at the project root and rendered as PNG diagrams showing the technologies in use.
📄️ errors
Get recent errors across all projects. Queries the error log aggregator and returns a unified view of failures from the last N hours.
📄️ exec
Execute a CLI command in a specific project's context. Runs an arbitrary command-line string inside a registered project's directory — useful for running project-specific tooling without cd-ing into the directory first.
📄️ run
Execute a named command on a registered project using the TOML manifest registry. Commands are defined in the project's manifest file. Arguments after the command name pass through to the underlying command.
📄️ route
Ticket routing commands. Analyzes a ticket description and suggests the appropriate project or team to handle it.
📄️ classify
Classify a ticket description as bug, feature, chore, or docs. Uses AI to determine ticket type and suggest routing. Outputs a classification with confidence and a suggested ADO work item type.
📄️ query
Query project state for interactive AI assistants. Returns structured JSON snapshots of project status, error logs, pipeline health, or system health. Designed to be piped into an AI context window.
📄️ game
Wasteology Trash Truck arcade game. A terminal arcade game where you drive a trash truck to collect all the trash while dodging raccoons. Three difficulty levels available.