{
  "markdown": "# learnship\n\n![learnship banner](assets/banner.png)\n\n<p align=\"center\">\n  <a href=\"https://github.com/FavioVazquez/learnship/actions/workflows/ci.yml\"><img src=\"https://github.com/FavioVazquez/learnship/actions/workflows/ci.yml/badge.svg?branch=main\" alt=\"CI\"></a>\n  <a href=\"https://github.com/FavioVazquez/learnship/releases/latest\"><img src=\"https://img.shields.io/github/v/release/FavioVazquez/learnship?color=3b82f6&label=release\" alt=\"Latest release\"></a>\n  <a href=\"https://faviovazquez.github.io/learnship/\"><img src=\"https://img.shields.io/badge/docs-faviovazquez.github.io%2Flearnship-2563eb\" alt=\"Documentation\"></a>\n  <a href=\"LICENSE\"><img src=\"https://img.shields.io/badge/License-MIT-22c55e.svg\" alt=\"License: MIT\"></a>\n  <a href=\"https://github.com/FavioVazquez/learnship/stargazers\"><img src=\"https://img.shields.io/github/stars/FavioVazquez/learnship?style=flat&color=f59e0b\" alt=\"Stars\"></a>\n  <a href=\"https://www.npmjs.com/package/learnship\"><img src=\"https://img.shields.io/npm/v/learnship?color=cb3837&label=npm\" alt=\"npm\"></a>\n  <img src=\"https://img.shields.io/badge/platforms-6-0ea5e9\" alt=\"6 platforms\">\n  <img src=\"https://img.shields.io/badge/workflows-57-3b82f6\" alt=\"57 workflows\">\n</p>\n\n<p align=\"center\">\n  <strong>Agentic engineering done right. Learn as you build. Build with intent.</strong><br>\n  <strong>A context engineering and spec-driven development system for Claude Code, Windsurf, Cursor, OpenCode, Gemini CLI, and Codex CLI.</strong>\n</p>\n\n<p align=\"center\">\n  <a href=\"#-install\">Install</a> ·\n  <a href=\"#-the-5-commands\">5 Commands</a> ·\n  <a href=\"#-the-phase-loop\">Phase Loop</a> ·\n  <a href=\"#-how-it-works\">How It Works</a> ·\n  <a href=\"#-workflow-reference-advanced\">All Workflows</a> ·\n  <a href=\"https://faviovazquez.github.io/learnship/\">Full Docs</a> ·\n  <a href=\"CHANGELOG.md\">Changelog</a>\n</p>\n\n---\n\n## ⚡ Install\n\n![Install learnship](assets/install.png)\n\n```bash\nnpx learnship\n```\n\n**Works on Mac, Windows, and Linux.** Requires Node.js ≥ 22 and Git. The installer auto-detects your platform.\n\n```bash\nnpx learnship --global   # all projects\nnpx learnship --local    # this project only\nnpx learnship --all --global  # all 6 platforms at once\n```\n\nThen open your AI agent and type:\n\n```\n/ls\n```\n\nThat's it. `/ls` tells you where you are, what to do next, and offers to run it.\n\n<details>\n<summary><strong>Platform-specific install + marketplace options</strong></summary>\n\n| Platform | Install command | Invoke commands as |\n|----------|----------------|-------------------|\n| **Windsurf** | `npx learnship --windsurf --global` | `/ls`, `/new-project` |\n| **Claude Code** | `npx learnship --claude --global` | `/learnship:ls`, `/learnship:new-project` |\n| **Cursor** | `/add-plugin learnship` *(marketplace)* | `@learnship` rules load automatically |\n| **OpenCode** | `npx learnship --opencode --global` | `/learnship-ls`, `/learnship-new-project` |\n| **Gemini CLI** | `npx learnship --gemini --global` | `/learnship:ls`, `/learnship:new-project` |\n| **Codex CLI** | `npx learnship --codex --global` | `$learnship-ls`, `$learnship-new-project` |\n\n**Via marketplace (no terminal required):**\n\n```bash\n# Claude Code — community marketplace\n/plugin marketplace add FavioVazquez/learnship-marketplace\n/plugin install learnship@learnship-marketplace\n\n# Cursor — after marketplace approval\n/add-plugin learnship\n\n# Gemini CLI — native extension\ngemini extensions install https://github.com/FavioVazquez/learnship\n```\n\n**Custom install directory:**\n\n```bash\nnpx learnship --claude --global --target /path/to/custom/dir\n```\n\n`--target` overrides the default platform directory. Works with install and uninstall on all 6 platforms.\n\nlearnship is published to npm — `npx learnship` pulls the latest release directly. No `github:` prefix, no clone needed. The same `bin/install.js` runs regardless of install method.\n\n</details>\n\n---\n\n## 🗺️ The 5 Commands\n\n![5 commands diagram](assets/quick-start-flow.png)\n\nlearnship has 57 workflows. You need five. Everything else surfaces naturally from `/ls`.\n\n| Command | What it does | When to use |\n|---------|-------------|-------------|\n| `/ls` | Show status, recent work, and next step (and offer to run it) | **Start every session here** |\n| `/next` | Read state and immediately run the right next workflow | When you just want to keep moving |\n| `/new-project` | Full init: questions → research → requirements → roadmap | Starting a new project |\n| `/quick \"...\"` | One-off task with atomic commits, no planning ceremony | Small fixes, experiments |\n| `/help` | All 57 workflows organized by category | Discovering capabilities |\n\n> **Tip:** `/ls` works for both new and returning users. No project? It explains learnship and offers `/new-project`. Returning? It shows your progress and suggests exactly what to do next.\n\n---\n\n## 🔄 The Phase Loop\n\n![Phase loop](assets/phase-loop.png)\n\nEvery feature ships through a 7-step loop:\n\n```mermaid\nflowchart LR\n    DP[\"/discuss-phase N<br/>Capture decisions\"]\n    PP[\"/plan-phase N<br/>Vertical slice plans\"]\n    EP[\"/execute-phase N<br/>Build + commit\"]\n    VW[\"/verify-work N<br/>UAT + diagnose\"]\n    RV[\"/review<br/>Two-pass review\"]\n    SH[\"/ship<br/>Test → PR\"]\n    CP[\"/compound<br/>Capture knowledge\"]\n\n    DP --> PP --> EP --> VW\n    VW --> RV --> SH --> CP\n    CP -->|\"next phase\"| DP\n    VW -->|\"all done\"| DONE[\"✓ /complete-milestone\"]\n```\n\n| Step | Command | What happens |\n|------|---------|-------------|\n| **1. Discuss** | `/discuss-phase N` | You and the agent align on implementation decisions before any code. Add `--deep` for extended deep questioning that walks every branch (v2.3.4) |\n| **2. Plan** | `/plan-phase N` | Agent researches the domain, creates vertical slice plans (tracer bullets), verifies them — including horizontal slice detection (v2.3.4) |\n| **3. Execute** | `/execute-phase N` | Plans run in dependency order, one atomic commit per task |\n| **4. Verify** | `/verify-work N` | You do UAT; agent diagnoses any gaps and creates fix plans |\n| **5. Review** | `/review` | Two-pass review: spec compliance check then 6-lens quality review (v2.4.0) |\n| **6. Ship** | `/ship` | Test → lint → commit → push → PR (v2.0) |\n| **7. Compound** | `/compound` | Capture what you learned as searchable documentation (v2.0) |\n\n**Just starting?** `/ls` or `/next` will route you into the right step automatically.\n\n---\n\n## 🏗️ How It Works\n\n![How it works](assets/how-it-works.png)\n\nThree integrated layers that reinforce each other:\n\n| Layer | What it does |\n|-------|-------------|\n| **Workflow Engine** | Spec-driven phases → context-engineered plans → wave-ordered execution → verified delivery |\n| **Learning Partner** | Neuroscience-backed checkpoints at every phase transition: retrieval, reflection, spacing, struggle |\n| **Design System** | 21 impeccable steering commands for production-grade UI: `/audit`, `/critique`, `/polish`, and more |\n\n---\n\n## 🌐 Platform Support\n\n![Platform comparison](assets/platform-comparison.png)\n\nEach platform gets the best experience it supports:\n\n| Feature | Windsurf | Claude Code | OpenCode | Gemini CLI | Codex CLI |\n|---------|----------|-------------|----------|------------|-----------|\n| Slash commands | ✓ | ✓ | ✓ | ✓ | `$skills` |\n| Real parallel subagents | — | ✓ | ✓ | ✓ | ✓ |\n| Parallel wave execution | — | ✓ | ✓ | ⚠️ experimental | ✓ |\n| Agent personas (17) | `model_decision` rules | `Task()` subagents | `Task()` subagents | `Task()` subagents | `Task()` subagents |\n| Interactive questions | `ask_user_question` | `AskUserQuestion` | `question` | `ask_user` | `request_user_input` |\n| Session hooks | — | ✓ | — | ✓ | — |\n| Skills (native `@invoke`) | ✓ | — | — | — | — |\n| Skills (context files) | ✓ | ✓ | ✓ | ✓ | ✓ |\n\n**Cursor** uses context injection via `cursor-rules/learnship.mdc` — it gets the full workflow library, design system, and learning partner, but does not appear in the feature matrix above because it has no first-class slash command or subagent API (rules load automatically as context).\n\n**Parallel subagents:** Claude Code, OpenCode, and Codex default to parallel execution — `execute-phase` spawns a dedicated executor per plan within a wave, each with its own 200k context budget. Up to 5 concurrent agents per wave. To run sequentially, set `\"parallelization\": { \"enabled\": false }` in `.planning/config.json`. Windsurf, Cursor, and Gemini CLI use sequential execution (no real subagent API on those platforms).\n\n---\n\n## What is learnship?\n\nlearnship is an **agent harness** — the scaffolding that makes your AI coding agent actually reliable across real projects.\n\nEvery serious AI coding tool converges on the same architecture: a simple execution loop wraps the model, and the **harness** decides what information reaches the model, when, and how. The model is interchangeable. The harness is the product.\n\nlearnship gives you that harness as a portable, open-source layer that adds:\n\n- **Persistent memory.** `/new-project` generates an `AGENTS.md` loaded automatically every session. No more repeating yourself.\n- **Structured process.** A repeatable phase loop with spec-driven plans, wave-ordered execution, and UAT-driven verification.\n- **Knowledge compounding.** `/compound` captures solved problems. `/review` runs two-pass code review (spec compliance then quality). `/ship` runs the full delivery pipeline.\n- **Security & recovery.** `/secure-phase` for STRIDE + OWASP Top 10 verification. `/forensics` for post-mortem. `/undo` for safe revert.\n- **Session intelligence.** Hooks, context profiles, interactive questions, agent delegation. ([v2.2 details →](#whats-new-in-v22))\n- **Built-in learning.** Neuroscience-backed checkpoints at every phase transition so you understand what you shipped.\n\n---\n\n## What problem does it solve?\n\nIf you've used AI coding assistants for more than a few sessions, you've hit this wall:\n\n> The agent forgets everything. Each session starts from scratch. Decisions get repeated. Code quality drifts. You ship fast but understand less.\n\nThis is a **harness problem**, not a model problem. The same model scores 42% with one scaffold and 78% with another. Same model. The only variable is the harness.\n\nlearnship solves this with **progressive disclosure** — context revealed incrementally, not dumped upfront. The right files, decisions, and phase context reach the agent exactly when needed.\n\n| Without learnship | With learnship |\n|-------------------|----------------|\n| Context resets every session | `AGENTS.md` loaded automatically every conversation |\n| Ad-hoc prompts, unpredictable results | Spec-driven plans, verifiable deliverables |\n| Architectural decisions get forgotten | `DECISIONS.md` tracked and honored by the agent |\n| Everything dumped into context at once | Phase-scoped context: only what this step needs |\n| You ship code you don't fully understand | Learning checkpoints build real understanding at every step |\n| UI looks generic, AI-generated | `impeccable` design system prevents AI aesthetic slop |\n\n---\n\n## Who is it for?\n\n**Anyone who wants to build and ship real products with AI agents** — founders, designers, researchers, makers, not just developers.\n\nIt's the right tool if:\n\n- You're **building a real project** and want the AI to stay aligned across sessions\n- You're **learning while building** and want to actually understand what gets shipped\n- You care about **code quality and UI quality** beyond \"it works\"\n- You want **parallel agent execution** — Claude Code, OpenCode, and Codex run it by default\n- You've felt the frustration of **context loss**: repeating yourself while the agent forgets\n\nIt's probably overkill for one-off scripts. Use `/quick` for that.\n\n---\n\n## 📚 Documentation\n\n**[faviovazquez.github.io/learnship](https://faviovazquez.github.io/learnship/)**\n\n- **[Getting Started](https://faviovazquez.github.io/learnship/getting-started/installation/)**: install, first project, the 5 commands\n- **[Platform Guide](https://faviovazquez.github.io/learnship/platform-guide/windsurf/)**: Windsurf, Claude Code, Cursor, OpenCode, Gemini CLI, Codex CLI\n- **[Core Concepts](https://faviovazquez.github.io/learnship/core-concepts/phase-loop/)**: phase loop, context engineering, planning artifacts\n- **[Skills](https://faviovazquez.github.io/learnship/skills/agentic-learning/)**: 11 `@agentic-learning` actions + 21 `impeccable` design commands\n- **[Workflow Reference](https://faviovazquez.github.io/learnship/workflow-reference/core/)**: all 57 workflows\n- **[Configuration](https://faviovazquez.github.io/learnship/configuration/)**: full schema, speed presets, parallelization\n\n---\n\n## 🆕 What's New\n\n### What's new in v2.4.0\n\nv2.4.0 adds spec compliance checking to `/review`, OWASP Top 10 coverage to `/secure-phase`, a numeric score to `/health`, and Playwright MCP smoke-test guidance to `/verify-work` and `/ship`:\n\n**Two-stage `/review`**: Pass 1 checks spec compliance — reads PLAN.md must-haves and classifies each as COVERED / PARTIAL / MISSING — before Pass 2 runs the existing 6-persona quality review. The spec compliance result appears in the report header. Use `--quality-only` to skip Pass 1 and run only the quality review.\n\n**OWASP Top 10 in `/secure-phase`**: The security-auditor agent now cross-maps STRIDE findings against OWASP Top 10 (A01–A10). Every SECURITY.md output includes an OWASP coverage table alongside the STRIDE analysis.\n\n**Numeric `/health` score**: The health check now outputs a 0–100 numeric score alongside the qualitative status. Starts at 100, deducts per issue found. Bands: HEALTHY (90–100), DEGRADED (70–89), BROKEN (0–69).\n\n**Playwright MCP guidance in `/verify-work` and `/ship`**: Optional live UI smoke-test sections activate when `@playwright/mcp` is configured. Supported on all 6 MCP-capable platforms (Claude Code, OpenCode, Cursor, Windsurf, Codex CLI, Gemini CLI). In `/verify-work`, walks the golden path using `mcp__playwright__*` tools. In `/ship`, runs a quick smoke test before creating the PR.\n\n### What's new in v2.3.4\n\nv2.3.4 adds two planning quality features:\n\n**Deep questioning mode** (`--deep` flag or `workflow.discuss_mode: \"deep\"` in config): Both `/discuss-phase` and `/new-project` now support extended questioning that walks every decision branch until shared understanding is reached. Each question includes a recommended answer. Standard mode (4 focused exchanges) remains the default.\n\n**Vertical slice planning** (enforced in `plan-phase`): Plans are now required to be tracer bullets — thin vertical slices through all integration layers (data → logic → API → UI → test) for one demoable user-facing behavior. The plan-checker flags any plan that covers only one architectural layer across all features. Single-layer phases (migrations, style passes) use `single_layer_justified: true` in the plan frontmatter.\n\n### What's new in v2.3\n\nv2.3 adds 5 new agent personas, Windsurf-native persona adoption via `model_decision` rules, and inline `<persona_context>` blocks across all 18 persona-aware workflows:\n\n**5 new agent personas**: `project-researcher` (domain ecosystem research for `/new-project`), `research-synthesizer` (synthesizes 4 research files into SUMMARY.md), `roadmapper` (creates phased roadmaps from requirements), `phase-researcher` (focused research for `/plan-phase` and `/research-phase`), `doc-verifier` (verifies docs match live code). Total agent pool: 17 specialist personas.\n\n**Windsurf `model_decision` rules**: Agent personas are now installed as `.windsurf/rules/learnship-{name}.md` with `trigger: model_decision` frontmatter. Windsurf's Cascade sees the rule description in every system prompt and reads the full persona when context is relevant — the native equivalent of Claude Code's subagent spawning.\n\n**Inline `<persona_context>` blocks**: All 18 workflows that reference agent personas now include inline persona instructions directly in the workflow text. This works on every platform — no special tool needed. Belt-and-suspenders with `@./agents/` file references and platform-native mechanisms.\n\n**Codex sandbox map**: All 17 agent personas now have per-agent sandbox modes (`read-only` for checkers/auditors, `workspace-write` for executors/planners).\n\n**Published agents synced**: The `agents/` directory now contains all 17 agents with proper frontmatter (`name:`, `description:`, `tools:`, `color:`) — in sync with the source `learnship/agents/` directory.\n\n### What's new in v2.2\n\nv2.2 adds session intelligence, structured interactivity, and research templates:\n\n**Session hooks** (Claude Code + Gemini CLI): 4 hooks installed via `settings.json` — statusline showing context usage, context monitor that warns before context runs out, prompt guard that scans `.planning/` writes for injection patterns, and session state that injects project orientation at startup.\n\n**Context profiles**: Set `\"context\": \"dev\"` (default), `\"research\"`, or `\"review\"` in config.json to control agent output style. Switch with `/settings`.\n\n**Interactive questions**: 14 workflows present decisions via your platform's native structured question tool — clickable cards on Claude Code, dropdowns on Windsurf, etc. `install.js` rewrites the tool name per platform automatically.\n\n**Agent persona delegation**: 18 workflows use inline `<persona_context>` blocks and `@./agents/` references for sequential persona adoption, with `Task()` subagent spawning when parallelization is enabled.\n\n**Research templates**: 5 structured fill-in-the-blanks templates (STACK.md, FEATURES.md, ARCHITECTURE.md, PITFALLS.md, SUMMARY.md) that prevent the AI from skipping file writes.\n\n**Upgrade safety**: SHA-256 file manifest after every install. Locally modified files detected and backed up before overwriting. Run `/reapply-patches` to restore customizations.\n\n### What's new in v2.1\n\nv2.1 adds 8 new workflows, 5 new references, 3 new templates, and 2 new agents:\n\n| Category | New workflows |\n|----------|--------------|\n| **Security** | `/secure-phase` — per-phase STRIDE threat-model security verification |\n| **Documentation** | `/docs-update` — generate and verify project docs against codebase |\n| **Recovery** | `/forensics` — post-mortem investigation · `/undo` — safe git revert |\n| **Session** | `/note` — zero-friction capture · `/session-report` — stakeholder summaries |\n| **Learning** | `/extract-learnings` — decisions, lessons, patterns, surprises · `/milestone-summary` — team onboarding |\n\nEnhanced: `/discuss-phase` (scope guardrails + domain probes + `--deep` extended questioning v2.3.4), `/new-project` (`--deep` extended questioning v2.3.4), `/plan-phase` (vertical slice tracer bullets + horizontal slice detection v2.3.4), `/execute-phase` (`--wave` flag + context scaling), `/quick` (`--research --validate --full` composable flags), `/ideate` (`--explore` Socratic mode).\n\n**Optional per-phase:** `/secure-phase N` (STRIDE security), `/extract-learnings N` (meta-knowledge).\n**Recovery:** `/forensics` (post-mortem), `/undo` (safe revert).\n\n---\n\n## ⚡ Agentic Engineering vs Vibe Coding\n\n![Vibe coding vs Agentic engineering](assets/vibe-vs-agentic.png)\n\n| | Vibe coding | Agentic engineering |\n|-|------------|--------------------|\n| **Context** | Resets every session | Engineered into every agent call |\n| **Decisions** | Implicit, forgotten | Tracked in `DECISIONS.md`, honored by the agent |\n| **Plans** | Ad-hoc prompts | Spec-driven, verifiable, wave-ordered |\n| **Outcome** | Code you shipped | Code you shipped **and understand** |\n\n---\n\n## 🧠 Context Engineering\n\n![Context engineering](assets/context-engineering.png)\n\nEvery agent invocation is loaded with structured context. Nothing is guessed:\n\n```mermaid\nflowchart LR\n    subgraph CONTEXT[\"Loaded into every agent call\"]\n        A[\"AGENTS.md<br/>Project soul + current phase\"]\n        B[\"REQUIREMENTS.md<br/>What we're building\"]\n        C[\"DECISIONS.md<br/>Every architectural choice\"]\n        D[\"Phase CONTEXT.md<br/>Implementation preferences\"]\n    end\n    CONTEXT --> AGENT[\"AI Agent\"]\n    AGENT --> P[\"Executable PLAN.md\"]\n    AGENT --> S[\"Commits + SUMMARY.md\"]\n```\n\n---\n\n## 🗂️ AGENTS.md: Persistent Project Memory\n\n![AGENTS.md](assets/agents-md.png)\n\n`/new-project` generates an `AGENTS.md` at your project root. On Windsurf, Claude Code, and Cursor it loads automatically every session. On other platforms, workflows reference it explicitly. Either way: the agent always knows the project, current phase, tech stack, and past decisions.\n\n```\nAGENTS.md                   ← your AI agent reads this every conversation\n├── Soul & Principles        # Pair-programmer framing, 10 working principles\n├── Platform Context         # Points to .planning/, explains the phase loop\n├── Current Phase            # Updated automatically by workflows\n├── Project Structure        # Filled during new-project from your answers\n├── Tech Stack               # Filled from research results\n└── Regressions              # Updated by /debug when bugs are fixed\n```\n\n---\n\n## 📖 Workflow Reference: Advanced\n\n> These are all 57 workflows. Most users discover them naturally from `/ls`. Scan this when you want to know if a specific capability exists.\n\n### Core Workflow\n\n| Workflow | Purpose | When to use |\n|----------|---------|-------------|\n| `/new-project` | Full init: questions → research → requirements → roadmap | Start of any new project |\n| `/discuss-phase [N]` | Capture implementation decisions before planning | Before every phase |\n| `/plan-phase [N]` | Research + create + verify plans | After discussing a phase |\n| `/execute-phase [N]` | Wave-ordered execution of all plans | After planning |\n| `/verify-work [N]` | Manual UAT with auto-diagnosis and fix planning. Optional Playwright MCP live UI smoke test when `@playwright/mcp` is configured. | After execution |\n| `/complete-milestone` | Archive milestone, tag release, prepare next | All phases verified |\n| `/audit-milestone` | Pre-release: requirement coverage, stub detection | Before completing milestone |\n| `/new-milestone [name]` | Start next version cycle | After completing a milestone |\n\n### Navigation\n\n| Workflow | Purpose | When to use |\n|----------|---------|-------------|\n| `/ls` | Status + next step + offer to run it | Start every session here |\n| `/next` | Auto-pilot: reads state and runs the right workflow | When you just want to keep moving |\n| `/progress` | Same as `/ls`: status overview with smart routing | \"Where am I?\" |\n| `/resume-work` | Restore full context from last session | Starting a new session |\n| `/pause-work` | Save handoff file mid-phase | Stopping mid-phase |\n| `/quick [description]` | Ad-hoc task with full guarantees | Bug fixes, small features |\n| `/help` | Show all available workflows | Quick command reference |\n\n### Phase Management\n\n| Workflow | Purpose | When to use |\n|----------|---------|-------------|\n| `/add-phase` | Append new phase to roadmap | Scope grows after planning |\n| `/insert-phase [N]` | Insert urgent work between phases | Urgent fix mid-milestone |\n| `/remove-phase [N]` | Remove future phase and renumber | Descoping a feature |\n| `/research-phase [N]` | Deep research only, no plans yet | Complex/unfamiliar domain |\n| `/list-phase-assumptions [N]` | Preview intended approach before planning | Validate direction |\n| `/plan-milestone-gaps` | Create phases for audit gaps | After audit finds missing items |\n\n### Brownfield, Discovery & Debugging\n\n| Workflow | Purpose | When to use |\n|----------|---------|-------------|\n| `/map-codebase` | Analyze existing codebase | Before `/new-project` on existing code |\n| `/discovery-phase [N]` | Map unfamiliar code area before planning | Entering complex/unfamiliar territory |\n| `/debug [description]` | Systematic triage → diagnose → fix | When something breaks |\n| `/diagnose-issues [N]` | Batch-diagnose all UAT issues by root cause | After verify-work finds multiple issues |\n| `/execute-plan [N] [id]` | Run a single plan in isolation | Re-running a failed plan |\n| `/add-todo [description]` | Capture an idea without breaking flow | Think of something mid-session |\n| `/check-todos` | Review and act on captured todos | Reviewing accumulated ideas |\n| `/add-tests` | Generate test coverage post-execution | After executing a phase |\n| `/validate-phase [N]` | Retroactive test coverage audit | After hotfixes or legacy phases |\n\n### Decision Intelligence\n\n| Workflow | Purpose | When to use |\n|----------|---------|-------------|\n| `/decision-log [description]` | Capture decision with context and alternatives | After any significant architectural choice |\n| `/knowledge-base` | Aggregate all decisions and lessons into one file | Before starting a new milestone |\n| `/knowledge-base search [query]` | Search the project knowledge base | When you need to recall why something was built a certain way |\n\n### Milestone Intelligence\n\n| Workflow | Purpose | When to use |\n|----------|---------|-------------|\n| `/discuss-milestone [version]` | Capture goals, anti-goals before planning | Before `/new-milestone` |\n| `/milestone-retrospective` | 5-question retrospective + spaced review | After `/complete-milestone` |\n| `/transition` | Write full handoff document for new session/collaborator | Before handing off or long break |\n\n### Compounding & Quality (v2.0)\n\n| Workflow | Purpose | When to use |\n|----------|---------|-------------|\n| `/compound` | Capture solved problem as searchable documentation | After `/debug`, `/verify-work`, or any aha moment |\n| `/review` | Two-pass review: spec compliance check then 6-persona quality review. `--quality-only` skips spec compliance. | After `/verify-work`, before shipping |\n| `/challenge` | Stress-test scope through product + engineering lenses | Before committing to a milestone or large feature |\n| `/ship` | Test → lint → commit → push → PR. Optional Playwright MCP smoke test before PR creation when `@playwright/mcp` is configured. | After review, ready to deploy |\n| `/ideate` | Codebase-grounded idea generation | Before `/discuss-milestone`, between milestones |\n| `/guard` | Safety mode: protect sensitive directories | Working on auth, payments, migrations |\n| `/sync-docs` | Detect stale documentation | Before `/complete-milestone`, after refactors |\n\n### Maintenance\n\n| Workflow | Purpose | When to use |\n|----------|---------|-------------|\n| `/settings` | Interactive config editor | Change mode, toggle agents |\n| `/set-profile [quality\\|balanced\\|budget]` | One-step model profile switch | Quick cost/quality adjustment |\n| `/health` | Project health check with numeric 0–100 score (HEALTHY ≥90, DEGRADED ≥70, BROKEN <70) | Stale files, missing artifacts |\n| `/cleanup` | Archive old artifacts | End of milestone |\n| `/update` | Update the platform itself | Check for new workflows |\n| `/reapply-patches` | Restore local edits after update | After `/update` if you had local changes |\n\n---\n\n## ⚙️ Configuration\n\nProject settings live in `.planning/config.json`. Set during `/new-project` or edit with `/settings`.\n\n### Full Schema\n\n```json\n{\n  \"mode\": \"interactive\",\n  \"granularity\": \"standard\",\n  \"model_profile\": \"balanced\",\n  \"learning_mode\": \"auto\",\n  \"context\": \"dev\",\n  \"test_first\": false,\n  \"planning\": {\n    \"commit_docs\": true,\n    \"commit_mode\": \"auto\",\n    \"search_gitignored\": false\n  },\n  \"workflow\": {\n    \"research\": true,\n    \"plan_check\": true,\n    \"verifier\": true,\n    \"validation\": true,\n    \"review\": true,\n    \"solutions_search\": true,\n    \"security_enforcement\": true,\n    \"discuss_mode\": \"discuss\",\n    \"tdd_mode\": false\n  },\n  \"parallelization\": {\n    \"enabled\": false,\n    \"plan_level\": true,\n    \"task_level\": false,\n    \"max_concurrent_agents\": 5,\n    \"min_plans_for_parallel\": 2\n  },\n  \"gates\": {\n    \"confirm_project\": true,\n    \"confirm_phases\": true,\n    \"confirm_roadmap\": true,\n    \"confirm_plan\": true,\n    \"execute_next_plan\": true,\n    \"issues_review\": true,\n    \"confirm_transition\": true\n  },\n  \"safety\": {\n    \"always_confirm_destructive\": true,\n    \"always_confirm_external_services\": true\n  },\n  \"review\": {\n    \"auto_after_verify\": false\n  },\n  \"ship\": {\n    \"auto_test\": true,\n    \"conventional_commits\": true,\n    \"pr_template\": true\n  },\n  \"hooks\": {\n    \"context_warnings\": true\n  },\n  \"git\": {\n    \"branching_strategy\": \"none\",\n    \"phase_branch_template\": \"phase-{phase}-{slug}\",\n    \"milestone_branch_template\": \"{milestone}-{slug}\"\n  }\n}\n```\n\n### Core Settings\n\n| Setting | Options | Default | What it controls |\n|---------|---------|---------|-----------------|\n| `mode` | `auto`, `interactive` | `auto` | `auto` auto-approves steps; `interactive` confirms at each decision |\n| `granularity` | `coarse`, `standard`, `fine` | `standard` | Phase size: 3-5 / 5-8 / 8-12 phases |\n| `model_profile` | `quality`, `balanced`, `budget` | `balanced` | Agent model tier (see table below) |\n| `learning_mode` | `auto`, `manual` | `auto` | `auto` offers learning at checkpoints; `manual` requires explicit invocation |\n| `context` | `dev`, `research`, `review` | `dev` | Output profile: `dev` (concise), `research` (detailed), `review` (audit-focused) |\n| `parallelization.enabled` | `true`, `false` | `true` (Claude Code, OpenCode, Codex) / `false` (others) | Parallel subagents per plan on supported platforms |\n| `test_first` | `true`, `false` | `false` | TDD mode: write failing test first, verify red, implement, verify green |\n| `planning.commit_mode` | `auto`, `manual` | `auto` | `auto` commits after each workflow step; `manual` skips all git commits |\n\n### Workflow Toggles\n\n| Setting | Default | What it controls |\n|---------|---------|------------------|\n| `workflow.research` | `true` | Domain research before planning each phase |\n| `workflow.plan_check` | `true` | Plan verification loop (up to 3 iterations), including vertical slice integrity check |\n| `workflow.verifier` | `true` | Post-execution verification against phase goals |\n| `workflow.validation` | `true` | Test coverage mapping during plan-phase |\n| `workflow.review` | `true` | Enable `/review` suggestions after `/verify-work` (v2.0) |\n| `workflow.solutions_search` | `true` | Search `.planning/solutions/` during `/plan-phase` (v2.0) |\n| `workflow.security_enforcement` | `true` | Per-phase STRIDE security verification via `/secure-phase` |\n| `workflow.discuss_mode` | `\"discuss\"` | Questioning depth: `\"discuss\"` (4 exchanges) or `\"deep\"` (extended, walks every branch) (v2.3.4) |\n| `workflow.tdd_mode` | `false` | Instruct planner to apply TDD task ordering to eligible tasks |\n\n### Review & Ship Settings\n\n| Setting | Default | What it controls |\n|---------|---------|------------------|\n| `review.auto_after_verify` | `false` | Auto-run `/review` after `/verify-work` passes |\n| `ship.auto_test` | `true` | Run test suite before shipping |\n| `ship.conventional_commits` | `true` | Use conventional commit format |\n| `ship.pr_template` | `true` | Auto-generate PR description |\n\n### Git Branching\n\n| `branching_strategy` | Creates branch | Best for |\n|---------------------|---------------|---------|\n| `none` | Never | Solo dev, simple projects |\n| `phase` | At each `execute-phase` | Code review per phase |\n| `milestone` | At first `execute-phase` | Release branches, PR per version |\n\n### Model Profiles\n\n| Agent | `quality` | `balanced` | `budget` |\n|-------|-----------|------------|----------|\n| Planner | large | large | medium |\n| Executor | large | medium | medium |\n| Phase Researcher | large | medium | small |\n| Debugger | large | medium | medium |\n| Verifier | medium | medium | small |\n| Plan Checker | medium | medium | small |\n| Solution Writer | medium | medium | small |\n| Code Reviewer | large | medium | medium |\n| Challenger | large | medium | medium |\n| Ideation Agent | large | medium | small |\n\n> **Platform note:** Tiers map to the best available model on your platform. On Claude Code: `large` = Opus, `medium` = Sonnet, `small` = Haiku. On Gemini CLI and Codex CLI the installer maps tiers to the best available model at install time. Windsurf, Cursor, and OpenCode use the platform default model — tiers signal intended task complexity.\n\n### Speed vs. Quality Presets\n\n| Scenario | `mode` | `granularity` | `model_profile` | Research | Plan Check | Verifier |\n|----------|--------|--------------|----------------|----------|------------|---------|\n| Prototyping | `auto` | `coarse` | `budget` | off | off | off |\n| Normal dev | `auto` | `standard` | `balanced` | on | on | on |\n| Production | `interactive` | `fine` | `quality` | on | on | on |\n\n---\n\n## 🧩 Learning Partner\n\nThe learning partner is woven into the platform, not bolted on. It fires at natural workflow transitions to build genuine understanding, not just fluent answers.\n\n### How it fires\n\n```\nlearning_mode: \"auto\"    → offered automatically at checkpoints (default)\nlearning_mode: \"manual\"  → only when you explicitly invoke @agentic-learning\n```\n\n### All 11 actions\n\n| Action | Trigger | What it does |\n|--------|---------|-------------|\n| `@agentic-learning learn [topic]` | Any time | Active retrieval: explain before seeing, then fill gaps |\n| `@agentic-learning quiz [topic]` | Any time | 3-5 questions, one at a time, formative feedback |\n| `@agentic-learning reflect` | After `execute-phase` | Three-question structured reflection: learned / goal / gaps |\n| `@agentic-learning space` | After `verify-work` | Schedule concepts for spaced review → writes `docs/revisit.md` |\n| `@agentic-learning brainstorm [topic]` | After `new-project` | Collaborative design dialogue before any code |\n| `@agentic-learning struggle [topic]` | During `quick` | Hint ladder: try first, reveal only when needed |\n| `@agentic-learning either-or` | After `discuss-phase` | Decision journal: paths considered, choice, rationale |\n| `@agentic-learning explain-first` | Any time | Oracy exercise: you explain, agent gives structured feedback |\n| `@agentic-learning explain [topic]` | Any time | Project comprehension log → writes `docs/project-knowledge.md` |\n| `@agentic-learning interleave` | Any time | Mixed retrieval across multiple topics |\n| `@agentic-learning cognitive-load [topic]` | After `plan-phase` | Decompose overwhelming scope into working-memory steps |\n\n**Core principle:** Fluent answers from an AI are not the same as learning. Every action makes you do the cognitive work, with support rather than shortcuts.\n\n### Skills across platforms\n\n| Platform | How `agentic-learning` works |\n|----------|-----------------------------|\n| **Windsurf** | Native skill: invoke with `@agentic-learning learn`, `@agentic-learning quiz`, etc. |\n| **Claude Code, OpenCode, Gemini CLI, Codex CLI** | Installed as a context file in `learnship/skills/agentic-learning/`. The AI reads and applies the techniques automatically. Reference it explicitly with `use the agentic-learning skill` or just work normally and it activates at checkpoints. |\n\n---\n\n## 🎨 Design System\n\nThe **impeccable** skill suite is always active as project context for any UI work. It provides design direction, anti-patterns, and 21 steering commands that prevent generic AI aesthetics. Based on [@pbakaus/impeccable](https://github.com/pbakaus/impeccable).\n\n### Commands\n\n| Command | What it does |\n|---------|-------------|\n| `/teach-impeccable` | One-time setup: gathers project design context and saves persistent guidelines |\n| `/audit` | Comprehensive audit: accessibility, performance, theming, responsive design |\n| `/critique` | UX critique: visual hierarchy, information architecture, emotional resonance |\n| `/polish` | Final quality pass: alignment, spacing, consistency before shipping |\n| `/normalize` | Normalize design to match your design system for consistency |\n| `/colorize` | Add strategic color to monochromatic or flat interfaces |\n| `/animate` | Add purposeful animations and micro-interactions |\n| `/bolder` | Amplify safe or boring designs for more visual impact |\n| `/quieter` | Tone down overly aggressive designs to reduce intensity and gain refinement |\n| `/distill` | Strip to essence: remove complexity, clarify what matters |\n| `/clarify` | Improve UX copy, error messages, microcopy, labels |\n| `/typeset` | Improve typography: font choices, hierarchy, sizing, weight, and readability |\n| `/arrange` | Improve layout, spacing, and visual rhythm; fix monotonous grids and weak hierarchy |\n| `/optimize` | Performance: loading speed, rendering, animations, bundle size |\n| `/harden` | Resilience: error handling, i18n, text overflow, edge cases |\n| `/delight` | Add moments of joy and personality that make interfaces memorable |\n| `/overdrive` | Push past conventional limits — shaders, spring physics, scroll-driven reveals |\n| `/extract` | Extract reusable components and design tokens into your design system |\n| `/adapt` | Adapt designs across screen sizes, devices, and contexts |\n| `/onboard` | Design onboarding flows, empty states, first-time user experiences |\n\n**The AI Slop Test:** If you showed the interface to someone and said \"AI made this\", would they believe you immediately? If yes, that's the problem. Use `/critique` to find out.\n\n### learnship integration\n\n**Automatic UI standards during `execute-phase`:** When a phase involves UI work, learnship detects it automatically and activates `@impeccable frontend-design` principles before any code is written. You'll see a banner announcing it. The agent then applies typography, color, layout, and component standards across every task in the phase — not as a post-hoc review but as an active constraint during execution.\n\n**Post-action milestone recommendation:** After any impeccable action produces recommendations, the agent suggests running `/new-milestone` to create a dedicated \"UI Polish\" milestone. This turns impeccable findings into versioned, traceable phases with plans and commits — so improvements don't disappear into chat history. Applying directly is always an option too.\n\n### Skills across platforms\n\n| Platform | How `impeccable` works |\n|----------|-----------------------|\n| **Windsurf** | Native skills: invoke each command directly with `/audit`, `/polish`, `/critique`, etc. |\n| **Claude Code, OpenCode, Gemini CLI, Codex CLI** | Installed as context files in `learnship/skills/impeccable/`. The AI reads design principles and anti-patterns automatically. Reference commands explicitly with `run the /audit impeccable skill` or just ask for UI work and it applies the standards. |\n\n---\n\n## 💡 Usage Examples\n\n### New greenfield project\n\n```\n/new-project              # Answer questions, configure, approve roadmap\n/discuss-phase 1          # Lock in your implementation preferences\n/plan-phase 1             # Research + plan + verify\n/execute-phase 1          # Wave-ordered execution\n/verify-work 1            # Manual UAT\n/review                   # two-pass review: spec compliance + quality (v2.4.0)\n/ship                     # v2.0: test → commit → push → PR\n/compound                 # v2.0: capture what you learned\n                          # Repeat for each phase\n/sync-docs                # v2.0: detect stale documentation\n/audit-milestone          # Check everything shipped\n/complete-milestone       # Archive, tag, done\n```\n\n### Existing codebase (brownfield)\n\n```\n/map-codebase             # Structured codebase analysis\n/new-project              # Questions focus on what you're ADDING\n# Normal phase workflow from here\n```\n\n### Quick bug fix\n\n```\n/quick \"Fix login button not responding on mobile Safari\"\n```\n\n### Quick with discussion + verification\n\n```\n/quick --discuss --full \"Add dark mode toggle\"\n```\n\n### Resuming after a break\n\n```\n/ls                       # See where you left off (offers to run next step)\n# or\n/next                     # Just pick up and go: auto-pilot\n# or\n/resume-work              # Full context restoration\n```\n\n### Scope change mid-milestone\n\n```\n/add-phase                # Append new phase to roadmap\n/insert-phase 3           # Insert urgent work between phases 3 and 4\n/remove-phase 7           # Descope phase 7 and renumber\n```\n\n### Preparing for release\n\n```\n/audit-milestone          # Check requirement coverage, detect stubs\n/plan-milestone-gaps      # If audit found gaps, create phases to close them\n/complete-milestone       # Archive, tag, done\n```\n\n### Debugging something broken\n\n```\n/debug \"Login flow fails after password reset\"\n```\n\n---\n\n## 🧭 Decision Intelligence\n\nEvery project accumulates decisions: architecture choices, library picks, scope trade-offs. The platform tracks them in a structured register so future sessions understand *why* the project is built the way it is.\n\n**`.planning/DECISIONS.md`** is the decision register:\n```markdown\n## DEC-001: Use Zustand over Redux\nDate: 2026-03-01 | Phase: 2 | Type: library\nContext: Needed client-side state for dashboard filters\nOptions: Zustand (simple, no boilerplate), Redux (complex, overkill for scope)\nChoice: Zustand\nRationale: 3x less boilerplate, sufficient for current data flow complexity\nConsequences: Locks React as UI framework; migration would require state rewrite\nStatus: active\n```\n\n**Populated automatically by:**\n- `discuss-phase` surfaces prior decisions before each phase discussion\n- `plan-phase` reads decisions before creating plans (never contradicts active ones)\n- `debug` puts architectural lessons from bugs into the register\n- `decision-log` manually captures any decision from any conversation\n\n**Queried by:**\n- `audit-milestone` checks decisions were honored in implementation\n- `knowledge-base` aggregates all decisions into a searchable `KNOWLEDGE.md`\n\n---\n\n## 📁 Planning Artifacts\n\nEvery project creates a structured `.planning/` directory:\n\n```\n.planning/\n├── config.json               # Workflow settings\n├── PROJECT.md                # Vision, requirements, key decisions\n├── REQUIREMENTS.md           # v1 requirements with REQ-IDs\n├── ROADMAP.md                # Phase breakdown with status tracking\n├── STATE.md                  # Current position, decisions, blockers\n├── DECISIONS.md              # Cross-phase decision register\n├── KNOWLEDGE.md              # Aggregated lessons (from knowledge-base)\n├── research/                 # Domain research from new-project\n│   ├── STACK.md\n│   ├── FEATURES.md\n│   ├── ARCHITECTURE.md\n│   ├── PITFALLS.md\n│   └── SUMMARY.md\n├── codebase/                 # Brownfield mapping (from map-codebase)\n│   ├── STACK.md\n│   ├── ARCHITECTURE.md\n│   ├── CONVENTIONS.md\n│   └── CONCERNS.md\n├── todos/\n│   ├── pending/              # Captured ideas awaiting work\n│   └── done/                 # Completed todos\n├── solutions/               # Knowledge compounding (from /compound) (v2.0)\n│   ├── auth/                # Solutions by category\n│   ├── performance/\n│   └── ...\n├── debug/                    # Active debug sessions\n│   └── resolved/             # Archived debug sessions\n├── quick/\n│   └── 001-slug/             # Quick task artifacts\n│       ├── 001-PLAN.md\n│       ├── 001-SUMMARY.md\n│       └── 001-VERIFICATION.md (if --full)\n└── phases/\n    └── 01-phase-name/\n        ├── 01-CONTEXT.md     # Your implementation preferences\n        ├── 01-DISCOVERY.md   # Unfamiliar area mapping (from discovery-phase)\n        ├── 01-RESEARCH.md    # Ecosystem research findings\n        ├── 01-VALIDATION.md  # Test coverage contract (from /validate-phase)\n        ├── 01-01-PLAN.md     # Executable plan (wave 1)\n        ├── 01-02-PLAN.md     # Executable plan (wave 1, independent)\n        ├── 01-01-SUMMARY.md  # Execution outcomes\n        ├── 01-UAT.md         # User acceptance test results\n        └── 01-VERIFICATION.md # Post-execution verification\n```\n\n---\n\n## 🔧 Troubleshooting\n\n### \"Project already initialized\"\n`/new-project` found `.planning/PROJECT.md` already exists. If you want to start over, delete `.planning/` first. To continue, use `/progress` or `/resume-work`.\n\n### Context degradation during long sessions\nStart each major workflow with a fresh context. The platform is designed around fresh context windows; every agent gets a clean slate. Use `/resume-work` or `/progress` to restore state after clearing.\n\n### Plans seem wrong or misaligned\nRun `/discuss-phase [N]` before planning. Most plan quality issues come from unresolved gray areas. Run `/list-phase-assumptions [N]` to see the intended approach before committing to a plan.\n\n### Execution produces stubs or incomplete code\nPlans with more than 3 tasks are too large for reliable single-context execution. Re-plan with smaller scope: `/plan-phase [N]` with finer granularity.\n\n### Lost track of where you are\nRun `/ls`. It reads all state files, shows your progress, and offers to run the next step.\n\n### Need to change something after execution\nUse `/quick` for targeted fixes, or `/verify-work` to systematically identify and fix issues through UAT. Do not re-run `/execute-phase` on a phase that already has summaries.\n\n### Costs running too high\nSwitch to budget profile via `/settings`. Disable research and plan-check for familiar domains. Use `granularity: \"coarse\"` for fewer, broader phases.\n\n### Working on a private/sensitive project\nSet `commit_docs: false` during `/new-project` or via `/settings`. Add `.planning/` to `.gitignore`. Planning artifacts stay local.\n\n### Something broke and I don't know why\nRun `/debug \"description of what's broken\"`. It runs triage → root cause diagnosis → fix planning with a persistent debug session.\n\n### Phase passed UAT but has known gaps\nRun `/audit-milestone` to surface all gaps, then `/plan-milestone-gaps` to create fix phases before release.\n\n---\n\n## 🚑 Recovery Quick Reference\n\n| Problem | Solution |\n|---------|----------|\n| Lost context / new session | `/ls` or `/next` |\n| Phase went wrong | `git revert` the phase commits, re-plan |\n| Need to change scope | `/add-phase`, `/insert-phase`, or `/remove-phase` |\n| Milestone audit found gaps | `/plan-milestone-gaps` |\n| Something broke | `/debug \"description\"` |\n| Quick targeted fix | `/quick` |\n| Plans don't match your vision | `/discuss-phase [N]` then re-plan |\n| Costs running high | `/settings` → budget profile, toggle agents off |\n\n---\n\n## 📂 Repository Structure\n\n```\nlearnship/\n├── .windsurf/\n│   ├── workflows/          # 57 workflows as slash commands\n│   ├── rules/              # 17 model_decision rules (agent personas for Windsurf)\n│   └── skills/\n│       ├── agentic-learning/   # Learning partner (SKILL.md + references), native on Windsurf + Claude Code\n│       └── impeccable/         # Design suite: 21 skills, native on Windsurf + Claude Code\n│           ├── frontend-design/ #   Base skill + 7 reference files (typography, color, motion…)\n│           ├── audit/           #   /audit\n│           ├── critique/        #   /critique\n│           ├── polish/          #   /polish\n│           └── …14 more/        #   /colorize /animate /bolder /quieter /distill /clarify…\n│                               # → on OpenCode/Gemini/Codex: both skills copied to learnship/skills/ as context files\n├── commands/               # 57 Claude Code-style slash command wrappers\n│   └── learnship/          # /learnship:ls, /learnship:new-project, etc.\n├── learnship/              # Payload installed into the target platform config dir\n│   ├── workflows/          # 57 workflow markdown files (the actual instructions)\n│   ├── references/         # Reference docs (questioning, verification, git, design, learning)\n│   └── templates/          # Document templates for .planning/ + AGENTS.md template\n├── agents/                 # 17 agent personas (planner, researcher, project-researcher, research-synthesizer, phase-researcher, roadmapper, executor, verifier, debugger, plan-checker, solution-writer, code-reviewer, challenger, ideation-agent, security-auditor, doc-writer, doc-verifier)\n├── assets/                 # Brand images (banner, explainers, diagrams)\n├── bin/\n│   └── install.js          # Multi-platform installer (Claude Code, OpenCode, Gemini CLI, Codex CLI, Windsurf)\n├── tests/\n│   └── run_all.sh               # 17 test suites, 1330+ checks across 6 platforms\n├── SKILL.md                # Meta-skill: platform context loaded by Cascade / AI agents\n├── install.sh              # Shell installer wrapper\n├── package.json            # npm package (npx learnship)\n├── CHANGELOG.md            # Version history\n└── CONTRIBUTING.md         # How to extend the platform\n```\n\n---\n\n## 🙏 Inspiration & Credits\n\n**learnship** builds on ideas and work from these open-source projects:\n\n- **[get-shit-done](https://github.com/gsd-build/get-shit-done)**: spec-driven development with structured workflows and planning artifacts — no sprint ceremonies, just build\n- **[agentic-learning](https://github.com/FavioVazquez/agentic-learning)**: neuroscience-backed learning techniques woven into the development cycle\n- **[impeccable](https://github.com/pbakaus/impeccable)**: frontend design quality system with auditing and refinement actions\n- **[compound-engineering](https://github.com/EveryInc/compound-engineering-plugin)**: the philosophy that each unit of engineering work should make subsequent units easier — compounding knowledge through structured review and documentation\n- **[superpowers](https://github.com/obra/superpowers)**: complete development workflow for coding agents with subagent-driven execution, TDD enforcement, and plan-based task dispatching\n- **[gstack](https://github.com/nichochar/gstack)**: builder-first engineering system with safety guards, shipping pipelines, multi-specialist review, and the \"Boil the Lake\" philosophy of AI-assisted completeness\n\nlearnship adapts, combines, and extends these into a unified, multi-platform system with integrated learning. All are used as inspiration and learnship is original work built on their shoulders.\n\n---\n\n## License\n\nMIT © [Favio Vazquez](https://github.com/FavioVazquez)\n",
  "bytes": 49769,
  "sha": "0d02fad48e9766fd0cb96405a52506c2a1881466c7d4a51a4dd4f251f07d9948",
  "repo_slug": "faviovazquez/learnship",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/plg_faviovazquez_learnship_66209aa0/readme"
}