Stats
Actions
Tags
From agency
Composable workflow commands for automating the complete software development lifecycle.
How this command is triggered — by the user, by Claude, or both
Slash command
/agency:READMEThe summary Claude sees in its command listing — used to decide when to auto-load this command
# Agency Commands Composable workflow commands for automating the complete software development lifecycle. ## Command Namespace All agency commands use the `agency:` prefix to avoid conflicts: - `/agency:work` instead of `/work` - `/agency:plan` instead of `/plan` ## Available Commands ### Core Workflow Commands #### `/agency:work` Work on any issue (auto-detects GitHub/Jira) **Usage**: **Flow**: Fetch → Plan → Review Plan → Implement → Test → Review → PR **Agents Used**: Orchestrator → Plan → Specialist Reviewer → Specialist Coder → Reality Checker --- #### `/agency:plan` Creat...
Composable workflow commands for automating the complete software development lifecycle.
All agency commands use the agency: prefix to avoid conflicts:
/agency:work instead of /work/agency:plan instead of /plan/agency:workWork on any issue (auto-detects GitHub/Jira)
Usage:
/agency:work 123 # Work on issue #123
/agency:work next # Work on next available issue
/agency:work https://... # Work on issue from URL
Flow: Fetch → Plan → Review Plan → Implement → Test → Review → PR
Agents Used: Orchestrator → Plan → Specialist Reviewer → Specialist Coder → Reality Checker
/agency:planCreate implementation plan only (no execution)
Usage:
/agency:plan 123 # Plan for issue #123
/agency:plan "add dark mode" # Plan from description
Flow: Analyze → Research → Plan → Review → Present
Agents Used: Orchestrator → Explore → Plan → Specialist Reviewer
/agency:implementImplement from existing plan
Usage:
/agency:implement <plan-file>
Flow: Load Plan → Implement → Test → Review → Commit
Agents Used: Orchestrator → Specialist Coder → Reality Checker
/agency:sprintImplement entire GitHub sprint
Usage:
/agency:sprint # Current sprint
/agency:sprint milestone-name # Specific milestone
Flow: Fetch Issues → Prioritize → Find Parallel → Execute Batches → Monitor
Agents Used: Orchestrator → Sprint Prioritizer → Multiple specialists in parallel
/agency:triageTriage GitHub issues
Usage:
/agency:triage # Triage open issues
/agency:triage label:bug # Triage specific label
Flow: Fetch Issues → Analyze → Categorize → Prioritize → Update Labels
Agents Used: Orchestrator → Senior Project Manager
/agency:parallelFind parallelizable GitHub issues
Usage:
/agency:parallel # Find all parallelizable work
/agency:parallel sprint # Only in current sprint
Flow: Fetch Issues → Analyze Dependencies → Group Independent → Report
Agents Used: Orchestrator → Senior Project Manager
/agency:worktreeCreate git worktree for issue
Usage:
/agency:worktree 123 # Create worktree for issue #123
Flow: Fetch Issue → Create Branch → Create Worktree → Setup Environment
/agency:sprintImplement entire Jira sprint
Usage:
/agency:sprint # Current sprint
/agency:sprint SPRINT-123 # Specific sprint
Flow: Fetch Issues → Prioritize → Find Parallel → Execute Batches → Monitor
/agency:triageTriage Jira issues
Usage:
/agency:triage # Triage backlog
/agency:triage project-key # Specific project
Flow: Fetch Issues → Analyze → Categorize → Prioritize → Update Jira
/agency:reviewComprehensive code review
Usage:
/agency:review # Review current changes
/agency:review PR-123 # Review specific PR
Flow: Security Review || Quality Review || Performance Review || Test Review → Synthesis
Agents Used: Legal Compliance Checker + Reality Checker + Performance Benchmarker + Test Results Analyzer (parallel)
/agency:testGenerate comprehensive tests
Usage:
/agency:test # Test current changes
/agency:test <file-path> # Test specific file
Flow: Analyze Code → Generate Unit Tests → Generate Integration Tests → Generate E2E Tests → Review Coverage
Agents Used: Orchestrator → Testing Specialist
/agency:securitySecurity audit workflow
Usage:
/agency:security # Audit current changes
/agency:security --full # Full codebase audit
Flow: Vulnerability Scan → Code Review → Dependency Check → Report
Agents Used: Legal Compliance Checker
/agency:optimizePerformance optimization workflow
Usage:
/agency:optimize # Optimize current code
/agency:optimize --profile # Profile first, then optimize
Flow: Profile → Identify Bottlenecks → Optimize → Benchmark → Verify
Agents Used: Performance Benchmarker → Backend/Frontend specialist
/agency:refactorRefactoring workflow
Usage:
/agency:refactor <file-or-pattern>
Flow: Analyze → Plan Refactoring → Execute → Test → Review
Agents Used: Senior Developer → Reality Checker
/agency:documentGenerate documentation
Usage:
/agency:document # Document current changes
/agency:document <file-path> # Document specific file
/agency:document --api # Generate API docs
Flow: Analyze Code → Generate Docs → Review → Format
/agency:adrCreate Architecture Decision Record
Usage:
/agency:adr "use PostgreSQL for persistence"
Flow: Context → Decision → Consequences → Document
/agency:work - Core workflow command/agency:plan - Planning only/agency:review - Code review/agency:sprint - Sprint automation/agency:triage - Issue triage/agency:parallel - Parallel work finder/agency:test - Test generation/agency:implement - From existing plan/agency:optimize - Performance optimization/agency:refactor - Refactoring workflow/agency:document - Documentation generation/agency:worktree - Worktree creation/agency:security - Security audit/agency:adr - ADR creationCommands should work together:
/agency:plan 123
# Review plan, make changes
/agency:implement plan.md
Commands should work without configuration:
/agency:work next # Auto-detects next issue
Commands support flags for advanced usage:
/agency:work 123 --auto-approve --model opus --no-tests
All commands follow similar structure:
Commands pause for user approval at key points:
See enhancements/outstanding-questions/ for detailed design questions about:
q1-command-priority-selection.md - Which commands to implement firstq2-interactive-vs-automated-commands.md - Interactive vs automated behaviorq3-command-argument-patterns.md - Argument format supportTo create a new command:
commands/agency-{name}.md---
description: Brief description of what this command does
argument-hint: Hint text shown to user
allowed-tools: [Read, Write, Edit, Bash, Task]
---
See existing commands for examples.
skills/agency-workflow-patterns/ - Orchestration patternsskills/github-workflow-best-practices/ - GitHub workflowsagents/orchestrator.md - Orchestrator agentenhancements/outstanding-questions/ - Design decisionsnpx claudepluginhub squirrelsoft-dev/squirrelsoft-marketplace --plugin agency/READMEDisplays documentation for DevTeam multi-agent slash commands: plan projects/sprints, implement autonomously, fix bugs/GitHub issues, review code, run tests, check status, manage git worktrees.
/agency-runOrchestrates a coordinated team of agents from .claude/agents/ to complete a task end-to-end. Supports FULL, QUICK, and HOTFIX sprint modes.
/orchestrateRuns a sequential multi-agent workflow for complex tasks — planning, implementation, review, and security audit — with structured handoffs and a final report.
/workflowOrchestrates a full multi-agent development pipeline: design, planning, plan review, implementation, and code review. Produces tested, reviewed code with commit and pipeline metrics.
/implementExecutes a single task through implementation, quality gates, and PR creation. Accepts plan documents, GitHub issues, or free-form descriptions.
/READMEDisplays documentation for all available commands, including format, validation, and discovery scripts. Also supports listing commands by keyword.