essentials-claude-code
starsforks
How to install essentials-claude-code?
gantisstorm/essentials-claude-code
$git clone https://github.com/GantisStorm/essentials-claude-codeAll-in-one workflow plugin—loops, swarms, and teams on Claude Code's Task System. All enforce exit criteria—swarm is faster with parallel queue execution, teams add contract-first coordination. Plan your way, execute your way. Optional: Beads for persistent memory, Ralph TUI for dashboard.
README.md
| 1 | <div align="center"> |
| 2 | |
| 3 | <img src="logo.png" alt="Essentials for Claude Code" width="450"/> |
| 4 | |
| 5 | # Essentials for Claude Code |
| 6 | |
| 7 | [](https://unlicense.org/) |
| 8 | [](https://claude.ai/code) |
| 9 | [](http://makeapullrequest.com) |
| 10 | |
| 11 | **Loops, swarms, and teams powered by Claude Code's built-in Task System.** |
| 12 | |
| 13 | Loop, swarm, and team are three execution modes. Loop runs sequentially. Swarm runs parallel subagents. Team spawns full Claude Code instances with shared contracts via [Agent Teams](https://www.anthropic.com/news/claude-opus-4-6). All use Claude's native task dependencies, `ctrl+t` progress, and automatic persistence. |
| 14 | |
| 15 | Plans define exit criteria. Loops run until tests pass. Done means actually done. |
| 16 | |
| 17 | </div> |
| 18 | |
| 19 | --- |
| 20 | |
| 21 | ## The Problem |
| 22 | |
| 23 | ``` |
| 24 | You: "Add authentication" |
| 25 | AI: *writes code* "Done!" |
| 26 | You: *runs tests* — 3 failing |
| 27 | You: "Fix these" |
| 28 | AI: "Fixed!" |
| 29 | You: *runs tests* — still failing |
| 30 | [repeat until you give up] |
| 31 | ``` |
| 32 | |
| 33 | ## The Solution |
| 34 | |
| 35 | ``` |
| 36 | # Option A: Discuss in chat, then execute from context |
| 37 | You: "I need to fix this auth bug..." [back and forth discussion] |
| 38 | You: /implement-loop fix the auth bug we discussed |
| 39 | AI: *implements, tests fail, fixes, tests fail, fixes...* |
| 40 | AI: "Exit criteria passed" ✓ |
| 41 | |
| 42 | # Option B: Create plan first, then execute |
| 43 | You: /plan-creator Add authentication |
| 44 | You: /plan-loop .claude/plans/auth-plan.md # Sequential |
| 45 | You: /plan-swarm .claude/plans/auth-plan.md # Parallel subagents |
| 46 | You: /plan-team .claude/plans/auth-plan.md # Agent Teams with contracts |
| 47 | AI: "Exit criteria passed" ✓ |
| 48 | ``` |
| 49 | |
| 50 | --- |
| 51 | |
| 52 | ## Quick Start |
| 53 | |
| 54 | ```bash |
| 55 | # Install |
| 56 | /plugin marketplace add GantisStorm/essentials-claude-code |
| 57 | /plugin install essentials@essentials-claude-code |
| 58 | mkdir -p .claude/plans .claude/prompts .claude/prd |
| 59 | |
| 60 | # Option A: From conversation (after discussing a bug/feature) |
| 61 | /implement-loop fix the auth bug we discussed # Sequential |
| 62 | /implement-swarm refactor the API handlers # Parallel subagents |
| 63 | /implement-team build the full-stack feature # Agent Teams with contracts |
| 64 | |
| 65 | # Option B: With plan file |
| 66 | /plan-creator Add user authentication with JWT |
| 67 | /plan-loop .claude/plans/user-auth-3k7f2-plan.md # Sequential |
| 68 | /plan-swarm .claude/plans/user-auth-3k7f2-plan.md # Parallel subagents |
| 69 | /plan-team .claude/plans/user-auth-3k7f2-plan.md # Agent Teams with contracts |
| 70 | |
| 71 | # Visual progress |
| 72 | ctrl+t # Toggle task tree view |
| 73 | ``` |
| 74 | |
| 75 | **Zero external dependencies.** All three modes enforce exit criteria. Swarm defaults to 3 concurrent workers. Team requires tmux and `CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1`. |
| 76 | |
| 77 | --- |
| 78 | |
| 79 | ## Powered by Claude Code's Task System |
| 80 | |
| 81 | Our loop, swarm, and team commands use Claude Code's **built-in Task Management System** (v2.1.19+). |
| 82 | |
| 83 | > **Ralph TUI and Beads integrate seamlessly.** Use [Ralph TUI](https://github.com/subsy/ralph-tui) for dashboard visualization. Use [Beads](https://github.com/steveyegge/beads) for persistent task tracking across sessions. |
| 84 | |
| 85 | ### What We Used Before |
| 86 | |
| 87 | The community built Ralph Wiggum loops with workarounds: |
| 88 | |
| 89 | - **Stop hooks**: Shell scripts (`.sh` files) that ran after each Claude response, grepping output for keywords like "complete" or "done" to decide whether to continue the loop |
| 90 | - **External plan files**: Markdown files tracking task state since Claude had no built-in persistence |
| 91 | - **TodoWrite**: Flat task lists with no dependency ordering — tasks could run out of order |
| 92 | - **Fresh sessions**: Starting new conversations to fight context rot, manually re-establishing state each time |
| 93 | |
| 94 | ### What We Use Now |
| 95 | |
| 96 | Claude Code v2.1.19+ provides native tools that replace all of this: |
| 97 | |
| 98 | | Old Workaround | Native Replacement | Why It's Better | |
| 99 | |----------------|-------------------|-----------------| |
| 100 | | Stop hooks (shell scripts) | `TaskUpdate({ status: "completed" })` | No external scripts, status is structured data | |
| 101 | | External plan.md for state | `~/.claude/tasks/` storage | Survives context compaction automatically | |
| 102 | | TodoWrite flat lists | `TaskUpdate({ addBlockedBy: [...] })` | Dependencies enforced — tasks can't run out of order | |
| 103 | | Manual session coordination | `CLAUDE_CODE_TASK_LIST_ID` env var | Same task list across sessions | |
| 104 | | Single agent | `TaskList` + parallel workers | Multiple agents coordinate via shared state | |
| 105 | |
| 106 | **The core loop is unchanged:** Plan → Implement → Verify → Loop if fail → Done when pass. |
| 107 | |
| 108 | ### What Tasks Do Well |
| 109 | |
| 110 | - **Dependency management**: Task #3 blocked by #1 and #2 literally cannot start until both complete |
| 111 | - **Visual progress**: Press `ctrl+t` to see live task tree with status |
| 112 | - **Parallel coordination**: Multiple workers share one task list, no conflicts |
| 113 | - **Persistence**: Survives context compaction, stored in `~/.claude/tasks/` |
| 114 | |
| 115 | ### What Tasks Don't Do |
| 116 | |
| 117 | `TaskList` shows ID, subject, status, and blockedBy — **but NOT description**. To see implementation d |
People who liked this also viewed
6 picksType
- claude-codeClaude Code is an agentic coding tool that lives in your terminal, understands your codebase, and helps you code faster by executing routine tasks, explaining complex code, and handling git workflows - all through natural language commands.WorkflowsJul 2026
- TrellisThe best agent harness.WorkflowsJul 2026
- reflyThe first open-source agent skills builder. Define skills by vibe workflow, run on Claude Code, Cursor, Codex & more. Build Clawdbot 🦞· APIs for Lovable · Bots for Slack & Lark/Feishu · Skills are infrastructure, not prompts.WorkflowsMar 2026
- pyspurA visual playground for agentic workflows: Iterate over your agents 10x fasterWorkflowsJun 2026
- claude-code🚀 Open source Claude Code CLI source code. Advanced AI Agent for developers. Includes TypeScript codebase for LLM tool-calling, agentic workflows, and terminal UI. Remember this is just the skeleton not the brain itself. Found by Chaofan Shou. WorkflowsApr 2026
- claude-codeClaude Code is an agentic coding tool that lives in your terminal, understands your codebase, and helps you code faster by executing routine tasks, explaining complex code, and handling git workflows - all through natural language commands.WorkflowsApr 2026
