A no-nonsense goal workflow engine for Claude Code and Codex: turn a plain-English goal into tracked steps with plain-language decisions and evidence-based done.
Show the active goal's status, proof, and what needs the user.
Start a new Goals-managed goal and begin its first phase.
Render a clean diagram of the goal's architecture or loop.
Turn accumulated, evidence-backed memory into a reviewable change set.
Advance the active goal — work the current phase and record evidence.
Generate a clean architecture or loop diagram for the active Goals goal as Mermaid or a valid .excalidraw file. Use when the user asks to visualize, diagram, or draw the goal's architecture map or the designed loop.
Show the architecture map of the active Goals goal — what is built, planned, blocked, deferred, or missing. Use inside a Goals project when you need the system shape, a diagram-style overview, or to check recorded changes against the goal's architecture.
Explain a project decision from the active Goals goal in plain language — what needs the user, the recommended reply, and what happens next. Use inside a Goals project when a technical decision, risk, or tradeoff needs a non-technical explanation, or when deciding whether a choice should interrupt the user.
Solve a problem or define/break down a goal using PACERS (Pause, Assess, Choose, Execute, Review, Systemize) and record the reasoning as a traceable building journey in the active Goals goal. Use inside a Goals project when defining a goal, breaking a phase into sub-problems, hunting assumptions before building, or whenever a problem is recurring, costly, unclear, or likely to create second-order effects.
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 no-nonsense goal workflow engine. It keeps your AI on a plan you can read, decide on, and trust.
Just say what you want — for example:
Anyone using AI to get real work done:
you say the goal
│
▼
Goals breaks it into clear steps ──▶ your AI does the next step
▲ │
│ you say yes ◀──── plain decision + proof it works
└───────── repeat until done — with a record of everything ◀─┘
Goals runs the workflow; your AI assistant (Claude Code, Codex, …) does the work. Goals is the part that keeps it organized, legible, and accountable.
1. Install Goals — one line. It installs everything it needs (including uv and
Python); nothing required up front.
macOS / Linux
curl -fsSL https://raw.githubusercontent.com/ShivamGupta42/goals/main/install.sh | sh
Windows (PowerShell)
irm https://raw.githubusercontent.com/ShivamGupta42/goals/main/install.ps1 | iex
2. Connect your AI assistant — one command, for Claude Code and Codex:
goals setup --agent both # or: --agent claude | --agent codex
That's it. Now just talk to it in Claude Code:
| You type | What happens |
|---|---|
/goals:create "build me a weight-loss tracking app" | Goals turns it into a tracked plan and starts step 1 |
/goals:next | Do the next step; Goals saves the proof and checks it off |
/goals:check | See where things stand and what (if anything) needs you |
Prefer the terminal? Use goals start "…", then goals next and goals check — see
The command set.
# needs uv (https://astral.sh/uv): curl -LsSf https://astral.sh/uv/install.sh | sh
uv tool install git+https://github.com/ShivamGupta42/goals.git
goals setup --agent both
AI agents wander. They skip steps, make quiet decisions, and after a while you've lost track of what they did. When they do turn to you to decide, it's hard — the choice comes wrapped in jargon you shouldn't have to decode. And once it's built, they often can't clearly tell you what they made or how.
Goals fixes that. Tell it what you want in plain English. It breaks that into a clear plan, keeps your AI on track step by step, puts every decision to you in plain words you can actually answer, and won't say "done" until there's proof. Works whether you write code or not.
Most people only need these:
| Command | What it does |
|---|---|
goals start "add login and payments to my site" | Turn a goal into a tracked plan and open a workspace for it |
goals next | Get the next step, ready to hand to your AI |
goals check | Plain-language status: progress, proof, and what needs you |
goals view | Open the dashboard — your goal at a glance, for humans |
goals loop | Design, check, and improve the workflow itself |
Under the hood, Goals is a small CLI + Claude Code / Codex plugin. It keeps goal state,
evidence, decisions, and an append-only history as plain files in your repo, plus a
portable spec any agent can pick up. On main it works in an isolated git worktree so
your checkout stays clean. goals check --json gives agents a machine-readable view.
Note: goals start runs in a git project (it makes a safe, isolated copy to work in).
Run goals --help for the full CLI, portability commands, and the visual loop builder.
Running a project with Goals? Add the badge to your README:
npx claudepluginhub shivamgupta42/goals --plugin goalsHarness-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 feature development workflow with specialized agents for codebase exploration, architecture design, and quality review
A growing collection of Claude-compatible academic workflow bundles. Covers scientific figures, manuscript writing and polishing, reviewer assessment, citation retrieval, data availability, paper reading, literature search, response letters, paper-to-PPTX conversion, and evidence-grounded Chinese invention patent drafting. Rules are organized as reusable skill folders with explicit workflows and quality checks.
Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques
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.
Tools to maintain and improve CLAUDE.md files - audit quality, capture session learnings, and keep project memory current.