A cross-tool agentic-coding workflow toolkit: 11 workflow skills (research / plan / implement / validate / commit / describe-pr / handle-pr-feedback / handoffs), an optional cross-vendor code-review skill, an LLM-as-judge evaluation skill, and a tracker abstraction (GitHub Issues, Azure DevOps work items, ...). Built on the Agent Skills open standard so the same files work in Claude Code and GitHub Copilot CLI.
Analyzes codebase implementation details. Call the codebase-analyzer agent when you need to find detailed information about specific components. As always, the more detailed your request prompt, the better! :)
Locates files, directories, and components relevant to a feature or task. Call `codebase-locator` with human language prompt describing what you're looking for. Basically a "Super Grep/Glob/LS tool" — Use it if you find yourself desiring to use one of these tools more than once.
codebase-pattern-finder is a useful subagent_type for finding similar implementations, usage examples, or existing patterns that can be modeled after. It will give you concrete code examples based on what you're looking for! It's sorta like codebase-locator, but it will not only tell you the location of files, it will also give you code details!
The research equivalent of codebase-analyzer. Use this subagent_type when wanting to deep dive on a research topic. Not commonly needed otherwise.
Discovers relevant documents in flow/ directory (We use this for all sorts of metadata storage!). This is really only relevant/needed when you're in a reseaching mood and need to figure out if we have random flow written down that are relevant to your current research task. Based on the name, I imagine you can guess this is the `flow` equivalent of `codebase-locator`
Stage and create git commits after showing a commit plan, with no AI attribution. Explicit-only; run only when the user invokes /commit or asks to commit, never autonomously.
Write a structured handoff document under flow/handoffs/ to transfer work to another session. Explicit-only; run only when invoked via /create-handoff or the user asks for a handoff.
Research the codebase and author a phased implementation plan under flow/plans/ from an issue / work item or task. Explicit workflow command; run only when invoked via /create-plan or explicitly asked, never autonomously.
Run an OPTIONAL, advisory cross-vendor code review of the current branch's diff vs main using a configurable second-opinion backend (Azure Foundry next-gen, GitHub Models, or Copilot CLI - all OAuth, all keyless). Explicit-only; run only when invoked via /cross-review or when the user explicitly asks for a cross-vendor review, never autonomously.
Generate a PR description from the repo template, save it under flow/prs/, and update the live PR. Explicit-only; run only when invoked via /describe-pr or the user asks to describe a PR.
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
A cross-tool agentic-coding workflow toolkit: 11 workflow skills (research → plan → implement → validate → review → ship), an optional cross-vendor code-review skill, and a tracker abstraction (GitHub Issues, Azure DevOps work items, GitLab issues, …) - all behind the Agent Skills open standard so the same files work in Claude Code and GitHub Copilot CLI.
Many agent failures are context failures: one mega-prompt tries to research, design, and code all at once, and quality degrades as the window fills (context rot). Flow stages the work - research, plan, implement, validate - and resets context between stages, so each step starts focused and leaves a durable artifact under flow/ for review. The same skills run in Claude Code and GitHub Copilot CLI, against GitHub Issues or Azure DevOps, with human checkpoints at the highest-leverage moments: the research and plan boundaries.
/plugin marketplace add corticalstack/flow
/plugin install flow@flow
/flow:init
/flow:init is an interactive bootstrap that creates AGENTS.md at your project root from Flow's template, creates .claude/tracker.json from the example, and optionally sets up cross-vendor review profiles.
gh skill install corticalstack/flow # interactive picker (recommended)
gh skill install corticalstack/flow .claude/skills/create-plan # one specific skill
Then drop in an AGENTS.md (see docs/AGENTS.md.template) and configure .claude/tracker.json (see .claude/tracker.example.json).
Workflow skills (11):
/flow:research-requirements, /flow:research-codebase - structured research phase/flow:create-plan, /flow:iterate-plan - implementation planning/flow:implement-plan, /flow:validate-plan - execution + quality gate/flow:commit - commit hygiene/flow:describe-pr, /flow:handle-pr-feedback - PR description + review-feedback loop/flow:create-handoff, /flow:resume-handoff - cross-session continuityOptional cross-vendor review:
/flow:cross-review - a configurable OAuth-only review against Azure AI Foundry, GitHub Models, or GitHub Copilot CLI. A model from a different vendor reads the diff vs main and surfaces high/critical findings - advisory only, never edits code. See docs/cross-vendor-review.md.Tracker abstraction:
bin/tracker (PATH-wrapper) + .claude/scripts/tracker.sh (the adapter)github backend wired (uses gh); azure-devops backend stub; none for tracker-less projects.claude/tracker.json. See docs/tracker-portability.md.Bootstrap helper:
/flow:init - creates AGENTS.md, .claude/tracker.json, and optional cross-review/profiles.json from templates./flow:research-requirements
↓
/flow:create-plan
↓
/flow:implement-plan ←─── iterate per phase
↓
/flow:validate-plan (quality gate)
↓
/flow:cross-review (optional, advisory)
↓
/flow:commit
↓
Push & PR
↓
/flow:describe-pr
↓
Review (human or @claude)
↓
/flow:handle-pr-feedback (if changes requested)
↓
Merge
Each stage drops a durable markdown artifact under flow/research/, flow/plans/, flow/prs/, or flow/reviews/, so the next agent (or future you) can pick up cleanly.
npx claudepluginhub corticalstack/flow --plugin flowEngineering skills for Claude Code: Terraform/IaC, FastAPI, MCP server development, modern Python, and scientific exploratory data analysis.
Thinking and reasoning skills for Claude Code: multi-perspective deliberation, scientific brainstorming, and context-engineering fundamentals.
Writing and content-creation skills for Claude Code: changelog generation, research-backed content writing, and AI-writing humanization.
Intelligent draw.io diagramming plugin with AI-powered diagram generation, multi-platform embedding (GitHub, Confluence, Azure DevOps, Notion, Teams, Harness), conditional formatting, live data binding, and MCP server integration for programmatic diagram creation and management.
Harness-native ECC operator layer - 67 agents, 271 skills, 92 legacy command shims, reusable hooks, rules, selective install profiles, and production-ready workflows for Claude Code, Codex, OpenCode, Cursor, and related agent harnesses
Comprehensive skill pack with 66 specialized skills for full-stack developers: 12 language experts (Python, TypeScript, Go, Rust, C++, Swift, Kotlin, C#, PHP, Java, SQL, JavaScript), 10 backend frameworks, 6 frontend/mobile, plus infrastructure, DevOps, security, and testing. Features progressive disclosure architecture for 50% faster loading.
Access thousands of AI prompts and skills directly in your AI coding assistant. Search prompts, discover skills, save your own, and improve prompts with AI.
Reliable automation, in-depth debugging, and performance analysis in Chrome using Chrome DevTools and Puppeteer
Comprehensive feature development workflow with specialized agents for codebase exploration, architecture design, and quality review