{ "markdown": "

\n \"Intent-Fluid\"\n

\n\n

\n Transforming human intent into autonomous execution.
\n Stop manual grinding, start directing intelligence.\n

\n\n

\n \"Platform\n \"Paradigm:\n \"Focus:\n

\n\n---\n\n## 🎬 See Surge In Action\n\n

\n \n

\n\n> **One PRD in → complete deliverables out.** Watch Surge autonomously iterate through Analyze → Research → Design → Implement → QA until quality converges.\n\n---\n\n## ⚡ The Difference\n\n

\n \"Without\n

\n\n---\n\n## 🌊 Why Surge?\n\n**🔄 Autonomous Iteration Engine**\nDirector Agent drives Analyze → Research → Design → Implement → QA in a closed loop. Not a one-shot generator — it iterates until quality converges. Auto-detects stagnation, oscillation, and Pareto frontiers to reach highest quality in minimum iterations.\n\n**👥 Expert Review Panel**\nAutomatically assembles 3-5 domain experts for parallel design review. Each expert evaluates independently — security, performance, maintainability — with veto power. Multi-perspective synthesis eliminates blind spots. High-risk decisions can optionally use expert-redteam-review for adversarial arbitration.\n\n**✅ Rigorous Quality Assurance**\nThree-tier acceptance criteria (L1→L2→L3) with progressive escalation. Output integrity validation auto-detects truncation and recovers. Optimization directives are closed-loop tracked — every improvement proposed is verified next round.\n\n**📋 PRD-to-Deliverable Pipeline**\nOne PRD in, complete deliverables out — code, documents, or strategy reports. Auto-analyzes requirement topology, negotiates acceptance criteria, orchestrates parallel subtasks. From fuzzy intent to structured output, fully autonomous.\n\n**🧠 Self-Evolving Process Memory**\nExtracts process experience after every iteration — ambiguities found, reusable components, rejected approaches, missing test cases. Persisted to memory files — gets smarter with every use. Retro phase generates rule update suggestions.\n\n---\n\n## 🧠 The Cognitive Pipeline\n\n

\n \"Surge\n

\n\nIntent-Fluid transforms intent into structured reality through an iterative, agentic pipeline. It doesn't generate once and stop — it loops through Analyze → Research → Design → Implement → QA until quality converges on your acceptance criteria. Each iteration extracts process memory, making the next run smarter.\n\n---\n\n## 📦 Quick Start\n\nSee the full [Installation Guide](docs/INSTALL.md) for platform-specific instructions.\n\n```bash\ngit clone https://github.com/carbonshow/intent-fluid.git\n```\n\n| Platform | Integration |\n|----------|-------------|\n| Claude Code | `/plugin marketplace add carbonshow/intent-fluid` |\n| Cursor | Add `https://github.com/carbonshow/intent-fluid` as a rule source or MCP server |\n| Gemini CLI | `gemini extensions install https://github.com/carbonshow/intent-fluid` |\n\n---\n\n## 🏗 Architecture\n\n

\n \"Intent-Fluid\n

\n\n### Available Skills\n\n| Skill | Description |\n|-------|-------------|\n| [ai-native-software-dev](skills/ai-native-software-dev/) | General AI-native software delivery workflow for brownfield systems, greenfield products, and authorized reference reconstruction. Combines an evidence-driven development kernel with risk tiers, conditional lifecycle modules, and extensible domain packs. |\n| [surge](skills/surge/) | Autonomous delivery system — iterative analyze/research/design/implement/QA cycles driven by a Director Agent. Provide a PRD to activate. |\n| [slidev](skills/slidev/) | Lightweight Markdown-based presentations — intelligently analyzes content, auto-selects layouts (text, diagrams, tables, images), generates Mermaid flowcharts, and applies adaptive density controls to prevent overflow. A faster, code-native alternative to heavy slide tools. |\n| [expert-redteam-review](skills/expert-redteam-review/) | Expert panel + red-team review workflow for complex decisions — scopes the decision, assembles domain experts, challenges assumptions, supports rebuttals, and produces judge arbitration with minimum validation actions. |\n| [domain-sensemaking](skills/domain-sensemaking/) | Structured research workflow for unfamiliar domains — frames the problem, generates and ranks exploration frontiers, builds concept/claim/evidence graphs, checks convergence, then synthesizes a traceable conclusion. Supports Learning, Research/Engineering, and Consulting/Decision modes. |\n\n### AI-Native Software Development\n\n`ai-native-software-dev` is an evidence-driven workflow for non-trivial software delivery. It uses one adaptable development kernel while providing dedicated paths for taking over large existing codebases, building products from zero, and independently reconstructing authorized product behavior. Risk tiers, conditional lifecycle modules, and composable domain packs keep the workflow lightweight for ordinary work while adding stronger controls for security-, data-, financial-, scientific-, or safety-sensitive systems.\n\nAsk your agent to use the skill, or invoke it explicitly where skill invocation is supported:\n\n```text\n$ai-native-software-dev Take over this repository and add CSV export without changing its authorization behavior.\n```\n\n### Creating a New Skill\n\n1. Read the [Skill Specification](docs/SKILL_SPEC.md)\n2. Copy the [Skill Template](docs/SKILL_TEMPLATE.md)\n3. Create your skill under `skills//`\n4. Validate: `bash scripts/validate-skill.sh skills/`\n\n---\n\n## 📄 License\n\nThis project is licensed under the MIT License — see the [LICENSE](LICENSE) file for details.\n\n---\n\n

\"Stay in flow. Let the intelligence follow your will.\"

\n", "bytes": 6629, "sha": "5e6e40636d545dc573f83ed9201ea2d94ab53b1353618cc5b3e834b0b6421933", "repo_slug": "carbonshow/intent-fluid", "fonte": "repo", "truncated": false, "api": "https://agentalog.com/api/listings/plg_carbonshow_intent_fluid_346db3ee/readme" }