workflow-automation

Workflow Automation Skill

Comprehensive guide for workflow automation, task management, and productivity optimization in software development. This skill combines CI/CD pipelines, git workflow management, scientific workflow tools, multi-agent orchestration, TDD workflows, and release automation.

When to Use This Skill

Use this skill when:

• CI/CD Pipelines: Creating, optimizing, or troubleshooting CI/CD workflows
• Git Workflows: Managing parallel development with worktrees and GitFlow
• Scientific Workflows: Choosing tools for computational workflows (joblib, Prefect, Parsl)
• Multi-Agent Orchestration: Coordinating multiple agents for complex tasks
• TDD/Development Cycles: Following RED-GREEN-REFACTOR with quality gates
• Code Review Automation: Implementing TRUST 5 framework validation
• Release Automation: Setting up semantic versioning and automated releases
• Task Management: Orchestrating complex dependencies and parallel execution

Quick Decision Guide

What are you trying to automate?

START: What type of workflow?

├── Build/Test/Deploy Pipeline
│   └── Use: CI/CD Pipelines → references/cicd/
│       ├── GitHub Actions → templates/github-actions/
│       └── GitLab CI → templates/gitlab-ci/

├── Parallel Feature Development
│   └── Use: Git Workflow Manager → scripts/git/
│       • create_worktree.sh
│       • GitFlow conventions

├── Scientific Computations
│   ├── Simple caching? → joblib (subskills/joblib.md)
│   ├── HPC cluster? → Parsl (subskills/parsl.md)
│   ├── Complex DAG? → Prefect (subskills/prefect.md)
│   └── Materials science? → quacc/FireWorks

├── Multi-Agent Coordination
│   └── Use: Alfred Workflow Orchestration
│       • WorkflowEngine patterns
│       • Context7 integration
│       • Error recovery

├── Development Workflow (TDD/Debug/Review)
│   └── Use: Development Workflow Specialist
│       • TDDManager → modules/tdd-context7.md
│       • AIDebugger → modules/ai-debugging.md
│       • AutomatedCodeReviewer → modules/automated-code-review.md

└── Automated Releases
    └── Use: Semantic Release → references/semantic-release/
        • Local-first releases
        • Conventional commits

Core Capabilities

1. CI/CD Pipelines

Design, optimize, and troubleshoot CI/CD pipelines across GitHub Actions and GitLab CI.

Key Features:

• Pipeline design patterns (fail-fast, parallelize, cache)
• DevSecOps security scanning (SAST, DAST, SCA)
• Deployment strategies (blue-green, canary, rolling)
• Performance optimization and troubleshooting

Quick Start:

# Basic pipeline structure
# 1. Fast feedback (lint, format) - <1 min
# 2. Unit tests - 1-5 min
# 3. Integration tests - 5-15 min
# 4. Build artifacts
# 5. Deploy (with approval gates)

Resources:

• references/cicd/best_practices.md - Pipeline design patterns
• references/cicd/security.md - Secrets management, OIDC
• references/cicd/devsecops.md - Security scanning guide
• templates/github-actions/ - GitHub Actions templates
• templates/gitlab-ci/ - GitLab CI templates

2. Git Workflow Management

Manage git worktrees following GitFlow conventions for parallel development.

Key Features:

• Worktree creation and management
• GitFlow branch conventions (feature/, fix/, hotfix/)
• Parallel feature development
• Clean branch organization

Quick Start:

# Create feature worktree
./scripts/git/create_worktree.sh feature email-notifications

# Result:
# - Branch: feature/email-notifications
# - Worktree: ../project-worktrees/feature/email-notifications/

# List all worktrees
./scripts/git/list_worktrees.sh

# Clean up merged worktrees
./scripts/git/cleanup_worktrees.sh --merged

Directory Structure:

project/                    ← Main repository (main branch)
project-worktrees/          ← Worktree parent
├── feature/
│   ├── email-notifications/
│   └── user-dashboard/
├── fix/
│   └── login-timeout/
└── hotfix/
    └── security-patch/

Resources:

• scripts/git/create_worktree.sh - Create worktree with GitFlow conventions
• scripts/git/list_worktrees.sh - List worktrees with status
• scripts/git/cleanup_worktrees.sh - Clean up stale worktrees
• references/gitflow-conventions.md - Complete GitFlow reference

3. Scientific Workflow Management

Choose the right tool for computational workflows - from simple joblib caching to complex orchestration.

Tool Selection:

Situation	Tool	Why
Cache function calls	joblib.Memory	Dead simple caching
10-100 tasks on laptop	joblib.Parallel	Built-in, no setup
100+ tasks on HPC	Parsl	HPC-aware, implicit parallelism
Complex DAG with retries	Prefect	Modern, Pythonic, great UI
Cloud-agnostic deployment	Covalent	Infrastructure abstraction
Materials DFT workflows	quacc	Pre-built recipes
Production materials	FireWorks	Battle-tested

Quick Start:

# Simple caching
from joblib import Memory
memory = Memory("./cache")

@memory.cache
def expensive_computation(x):
    return result

# Parallel execution
from joblib import Parallel, delayed
results = Parallel(n_jobs=4)(
    delayed(compute)(i) for i in range(100)
)

Resources:

• subskills/joblib.md - Caching and parallelism
• subskills/prefect.md - Modern Python orchestration
• subskills/parsl.md - HPC scientific computing
• subskills/covalent.md - Cloud/quantum workflows
• subskills/fireworks.md - Materials production
• subskills/quacc.md - Materials high-throughput

4. Multi-Agent Workflow Orchestration

Coordinate multiple agents for complex development workflows with Context7 integration.

Key Features:

• Multi-agent coordination
• Task scheduling with priorities
• Error handling and recovery
• Performance monitoring

Quick Start:

from alfred_workflow import WorkflowEngine, Agent

engine = WorkflowEngine()
spec_agent = Agent("spec-builder", domain="requirements")
impl_agent = Agent("tdd-implementer", domain="development")
test_agent = Agent("quality-gate", domain="testing")

workflow = engine.create_workflow("feature_development")
workflow.add_stage("specification", spec_agent)
workflow.add_stage("implementation", impl_agent, depends_on=["specification"])
workflow.add_stage("testing", test_agent, depends_on=["implementation"])

result = engine.execute(workflow, input_data={"feature": "user auth"})

Workflow Templates:

• Feature Development: spec → implementation → testing
• Bug Fix: diagnosis → fix → validation
• Code Review: analysis → feedback → approval

Resources:

• references/alfred-workflow-patterns.md - Orchestration patterns
• examples/enterprise-testing-workflow.py - Complete workflow example

5. Development Workflow (TDD/Debug/Review)

Implement TDD cycles, AI-powered debugging, and automated code review with TRUST 5 framework.

Key Components:

TDDManager - Test-driven development cycle:

from workflow import TDDManager, TestSpecification

tdd = TDDManager(project_path, context7_client)

# RED: Generate failing test
# GREEN: Implement minimum code
# REFACTOR: Optimize with patterns
cycle_results = await tdd.run_full_tdd_cycle(
    specification=test_spec,
    target_function="authenticate_user"
)

AIDebugger - Intelligent error analysis:

from workflow import AIDebugger

debugger = AIDebugger(context7_client)
analysis = await debugger.debug_with_context7_patterns(
    exception=e,
    context={"file": "app.py", "function": "main"},
    project_path="/project/src"
)
# Returns: root cause, solutions, code suggestions

AutomatedCodeReviewer - TRUST 5 validation:

from workflow import AutomatedCodeReviewer

reviewer = AutomatedCodeReviewer(context7_client)
review = await reviewer.review_codebase(
    project_path="/project/src",
    changed_files=["src/auth/service.py"]
)
# TRUST 5: Test-first, Readable, Unified, Secured, Trackable

Resources:

• modules/tdd-context7.md - TDD with Context7
• modules/ai-debugging.md - AI-powered debugging
• modules/automated-code-review.md - TRUST 5 review
• modules/smart-refactoring.md - Technical debt analysis
• modules/performance-optimization.md - Profiling and optimization

6. Release Automation (Semantic Release)

Automate versioning and releases using semantic-release (Node.js) for any language.

Key Features:

• Local-first releases (instant feedback vs 2-5 min CI wait)
• Conventional commits for automatic version bumps
• GitHub releases and changelog generation
• Multi-account authentication support

Quick Start:

# Dry-run first
GITHUB_TOKEN=$(gh auth token) npx semantic-release --no-ci --dry-run

# Create release
GITHUB_TOKEN=$(gh auth token) npx semantic-release --no-ci

Version Bump Rules:

• feat: → MINOR (0.1.0 → 0.2.0)
• fix: → PATCH (0.1.0 → 0.1.1)
• BREAKING CHANGE: → MAJOR (0.1.0 → 1.0.0)

Resources:

• references/semantic-release/ - Complete release workflow
• references/semantic-release/local-release-workflow.md - Local release guide
• references/semantic-release/authentication.md - SSH and gh CLI setup

7. FABER Workflow State Management

Manage workflow state with FABER (Frame, Architect, Build, Evaluate, Release) methodology.

Key Operations:

# Load configuration
./scripts/faber/config-loader.sh

# Read workflow state
./scripts/faber/state-read.sh ".faber/state.json"

# Update phase state
./scripts/faber/state-update-phase.sh frame completed '{"work_type": "feature"}'

# Post status card
./scripts/faber/status-card-post.sh abc12345 123 evaluate "Build is green"

Resources:

• scripts/faber/ - FABER workflow framework (29 scripts)
• templates/status-card.template.md - Status card template

Quality Metrics

Metric	Threshold	Description
Test Coverage	>= 85%	Minimum code coverage
TRUST Score	>= 0.85	5 quality criteria combined
Critical Issues	0	No critical security/bugs
Performance Regression	< 10%	Max allowed degradation
Response Time	< 100ms	API response target

Best Practices

CI/CD Pipelines

• Fail fast: Run cheap validation first
• Parallelize: Remove unnecessary job dependencies
• Cache dependencies: 50-90% faster builds
• Use OIDC: No static credentials
• Pin actions: Use commit SHAs

Git Workflows

• One feature per worktree
• Descriptive branch names
• Commit before switching
• Push regularly for backup
• Clean up after merge

Development Workflows

• Write tests before implementation (TDD)
• Maintain 85%+ coverage
• Use meaningful test names
• Mock external dependencies
• Profile before optimizing

Release Automation

• Use conventional commits
• Run dry-run before release
• Verify account alignment
• Local releases for speed
• CI backup optional

Resources Directory Structure

workflow-automation/
├── SKILL.md                     # This file
├── modules/
│   ├── ai-debugging.md          # AIDebugger class
│   ├── automated-code-review.md # TRUST 5 framework
│   ├── performance-optimization.md
│   ├── smart-refactoring.md
│   └── tdd-context7.md          # TDDManager class
├── references/
│   ├── cicd/
│   │   ├── best_practices.md
│   │   ├── devsecops.md
│   │   ├── optimization.md
│   │   ├── security.md
│   │   └── troubleshooting.md
│   ├── gitflow-conventions.md
│   ├── alfred-workflow-patterns.md
│   └── semantic-release/
├── scripts/
│   ├── git/                     # Git worktree scripts
│   ├── faber/                   # FABER workflow framework (29 scripts)
│   ├── cicd/                    # Pipeline analysis scripts
│   └── workflow/                # Workflow utilities
├── subskills/
│   ├── joblib.md
│   ├── prefect.md
│   ├── parsl.md
│   ├── covalent.md
│   ├── fireworks.md
│   └── quacc.md
├── templates/
│   ├── github-actions/
│   │   ├── node-ci.yml
│   │   ├── python-ci.yml
│   │   ├── go-ci.yml
│   │   ├── docker-build.yml
│   │   └── security-scan.yml
│   └── gitlab-ci/
│       ├── node-ci.yml
│       ├── python-ci.yml
│       ├── go-ci.yml
│       ├── docker-build.yml
│       └── security-scan.yml
├── examples/
│   └── enterprise-testing-workflow.py
└── docs/
    └── quick-reference.md

See Also

• references/cicd/ - CI/CD deep-dive documentation
• modules/ - Development workflow modules
• subskills/ - Scientific workflow tool guides
• templates/ - Pipeline starter templates
• scripts/ - Automation scripts

---

Version: 1.0.0 Last Updated: 2025-01-18 Sources: development-workflow-specialist, git-workflow-manager, scientific-workflow-management, alfred-workflow-orchestration, faber-core, cicd-pipelines, semantic-release