From dev-standards
Execute the standards orchestrator with phase sequencing, checkpoints, and handoffs. Run development standards across domains (logging, error, type, lint, test, validation, git, housekeeping, naming) with enforced workflow phases: Design -> Validate Design -> Build -> Test -> Validate.
How this skill is triggered — by the user, by Claude, or both
Slash command
/dev-standards:orchestrateThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Execute the standards orchestrator runtime to apply development standards systematically.
Execute the standards orchestrator runtime to apply development standards systematically.
/orchestrate domain=<domain> [phase=<phase>]
| Parameter | Required | Description |
|---|---|---|
domain | Yes | Domain to orchestrate. Options: logging, error, type, lint, test, validation, git, housekeeping, naming, all |
phase | No | Specific phase to start from. Options: design, validate-design, build, test, validate |
# Run all phases for git standards
/orchestrate domain=git
# Run all phases for all domains
/orchestrate domain=all
# Run only design phase for logging
/orchestrate domain=logging phase=design
The orchestrator executes these phases in sequence:
After design and build phases, execution pauses for user approval:
approve, yes, proceed, or lgtm to continuereject or rollback to stop executionEach domain has a dedicated agent with specific expertise:
| Domain | Agent | Expertise |
|---|---|---|
logging | @operate-logging-reviewer | Structured logging, log levels, debug mode |
error | @operate-error-handling-reviewer | Try/catch patterns, error types, stack traces |
type | @validation-type-safety-reviewer | JSDoc types, TypeScript, mypy |
lint | @validation-lint-reviewer | ESLint, ruff, auto-fix, rule config |
test | @testing-strategy-reviewer | Unit/integration tests, coverage, mocking |
validation | @validation-standards-reviewer | Input validation, sanitization, security |
git | @operate-git-workflow-reviewer | Conventional commits, branch naming |
housekeeping | @maintain-housekeeping-reviewer | Project layout, temp dirs, clutter |
naming | @meta-naming-standards-reviewer | File/function naming conventions |
When one agent's work requires another agent (e.g., new error handling needs logging), handoffs are tracked:
@operate-error-handling-reviewer (build) -> @operate-logging-reviewer: Add logging to catch blocks
@operate-logging-reviewer (build) -> @testing-strategy-reviewer: Add test coverage
The orchestrator can also be controlled via CLI:
# Initialize
node lib/orchestrator/cli.js init domain=git
# Check status
node lib/orchestrator/cli.js status
# Advance to next phase
node lib/orchestrator/cli.js advance
# Handle checkpoint
node lib/orchestrator/cli.js checkpoint approve
node lib/orchestrator/cli.js checkpoint reject "Need more tests"
# Get current agent prompt
node lib/orchestrator/cli.js prompt
# View progress
node lib/orchestrator/cli.js progress
# Manage handoffs
node lib/orchestrator/cli.js handoff status
node lib/orchestrator/cli.js handoff register '{"to":"test-standards","reason":"Add tests"}'
# Reset
node lib/orchestrator/cli.js reset
Orchestrator state is saved to tmp/.orchestrator-state.json and persists across sessions:
Use /orchestrate when you want to:
User: /orchestrate domain=git
Claude: Initializing orchestrator for domain: git
Phase: design
Loading @operate-git-workflow-reviewer context...
[Executes design phase, proposes changes]
## Checkpoint: design Complete
**Agent**: @operate-git-workflow-reviewer
**Phase**: design
**Status**: Awaiting user approval
### Changes Proposed
- Add conventional commit validation
- Add pre-commit hook for message format
- Block AI co-authoring in commits
User Action Required:
- [ ] Approve and proceed
- [ ] Request modifications
- [ ] Reject and rollback
User: approve
Claude: Checkpoint approved. Advancing to: validate-design
...
The orchestrator integrates with plugin hooks:
This ensures the workflow cannot be bypassed accidentally.
npx claudepluginhub myenquiringmind/dev-standards-plugin --plugin dev-standardsGuides creation, editing, and verification of skills for AI coding agents using test-driven development with subagent scenarios. Use when authoring or debugging skills.