{"_id":"@aiedx/firstbreakai","_rev":"3-c6a70e1a28798b981d129e307c3f53fb","name":"@aiedx/firstbreakai","dist-tags":{"latest":"0.3.0"},"versions":{"0.1.0":{"name":"@aiedx/firstbreakai","version":"0.1.0","keywords":["ai","cohort","llm","machine-learning","huggingface","quarto","first-break-ai","mcp","inference","training"],"author":{"url":"https://cohort.bubblnet.com","name":"First Break AI","email":"contact@bubblspace.com"},"license":"MIT","_id":"@aiedx/firstbreakai@0.1.0","maintainers":[{"name":"adssidhu86","email":"aiedx.org@gmail.com"}],"homepage":"https://cohort.bubblnet.com","bugs":{"url":"https://github.com/thefirehacker/firstbreakai/issues"},"bin":{"firstbreakai":"bin/firstbreakai.js"},"dist":{"shasum":"e395611172dfea6908981999b28b49c5b2d16331","tarball":"https://registry.npmjs.org/@aiedx/firstbreakai/-/firstbreakai-0.1.0.tgz","fileCount":17,"integrity":"sha512-tL4wj9afRhz2MsDXm15XhGs3q4TvwzMIcLp4FWJj8lbBS5+qOcsi7ppsfXXLlPtH/P2DWECKdMKXQ+jfJbGGqg==","signatures":[{"sig":"MEUCIQDSPGReCLKqhwpy1Bq4jDpDqmQu6uqRX2N6mY6/LfoEgAIgWQUIyjRG9m+1aIO0d08MA6oJiGdzrVTM+CCaYk5zqXw=","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":33591},"type":"module","engines":{"node":">=18"},"gitHead":"0707d02c3004dfa96f08ff2f725f083c599521a8","_npmUser":{"name":"adssidhu86","email":"aiedx.org@gmail.com"},"repository":{"url":"git+https://github.com/thefirehacker/firstbreakai.git","type":"git"},"_npmVersion":"10.7.0","description":"CLI and local MCP server for the First Break AI cohort — check your environment, track progress, validate exercises, and get help.","directories":{},"_nodeVersion":"22.2.0","dependencies":{"open":"^10.1.0","chalk":"^5.3.0"},"_hasShrinkwrap":false,"_npmOperationalInternal":{"tmp":"tmp/firstbreakai_0.1.0_1780502088958_0.8963818252733828","host":"s3://npm-registry-packages-npm-production"}},"0.2.0":{"name":"@aiedx/firstbreakai","version":"0.2.0","keywords":["ai","cohort","llm","machine-learning","huggingface","quarto","first-break-ai","mcp","inference","training"],"author":{"url":"https://cohort.bubblnet.com","name":"First Break AI","email":"contact@bubblspace.com"},"license":"MIT","_id":"@aiedx/firstbreakai@0.2.0","maintainers":[{"name":"adssidhu86","email":"aiedx.org@gmail.com"}],"homepage":"https://cohort.bubblnet.com","bugs":{"url":"https://github.com/thefirehacker/firstbreakai/issues"},"bin":{"firstbreakai":"bin/firstbreakai.js"},"dist":{"shasum":"e68afe01b62ea4057dc4a081a0e6530a4509c424","tarball":"https://registry.npmjs.org/@aiedx/firstbreakai/-/firstbreakai-0.2.0.tgz","fileCount":19,"integrity":"sha512-lVKrr/q8fHmknLw4DXJHaIbB0xlniTYoESAzdW3xcbY8d98hp3hp4S13WWmH8qjjv3cYcJdvZUuS6EG4h44PpA==","signatures":[{"sig":"MEMCIAJ1sGPcb8tXsEEdLrN2tf2WLwHfA0YVgSj1tdPhReKTAh9WzgpOER1TaoQUx1er80XEUkvGH7Hgz51EIATy+gKJ","keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U"}],"unpackedSize":42699},"type":"module","engines":{"node":">=18"},"gitHead":"660e107ee779e5b9c0247441e66b3c554964d378","_npmUser":{"name":"adssidhu86","email":"aiedx.org@gmail.com"},"repository":{"url":"git+https://github.com/thefirehacker/firstbreakai.git","type":"git"},"_npmVersion":"10.7.0","description":"CLI and local MCP server for the First Break AI cohort — check your environment, track progress, validate exercises, and get help.","directories":{},"_nodeVersion":"22.2.0","dependencies":{"open":"^10.1.0","chalk":"^5.3.0"},"_hasShrinkwrap":false,"_npmOperationalInternal":{"tmp":"tmp/firstbreakai_0.2.0_1780503788652_0.5246667974094066","host":"s3://npm-registry-packages-npm-production"}},"0.3.0":{"name":"@aiedx/firstbreakai","version":"0.3.0","description":"CLI + local MCP server for First Break AI — the first fully agentic AI cohort, where AI agents check your environment, validate your code, track progress, and answer questions. Powered by FetchLens.ai.","homepage":"https://cohort.bubblnet.com","repository":{"type":"git","url":"git+https://github.com/thefirehacker/firstbreakai.git"},"bugs":{"url":"https://github.com/thefirehacker/firstbreakai/issues"},"keywords":["ai","cohort","llm","machine-learning","huggingface","quarto","first-break-ai","mcp","inference","training","agentic-ai","ai-agents","ai-agent","agentic","model-context-protocol","mcp-server","claude","cursor","codex","ai-tutor"],"license":"MIT","author":{"name":"First Break AI","email":"contact@bubblspace.com","url":"https://cohort.bubblnet.com"},"type":"module","bin":{"firstbreakai":"bin/firstbreakai.js"},"engines":{"node":">=18"},"dependencies":{"chalk":"^5.3.0","open":"^10.1.0"},"_id":"@aiedx/firstbreakai@0.3.0","gitHead":"07edda0fd46daab7ccf0966069aea586a2feb3df","_nodeVersion":"22.2.0","_npmVersion":"10.7.0","dist":{"integrity":"sha512-ugTDw6cXNGLHhp74rodNH4WZB3TLGREzbwZQd0Hgh4/GKljeLDCJ5aCjm3oj0gTpkq8m3Vq7pX/da1d2tGC3wA==","shasum":"40e3f87601472109ec43d2e086c061a95e8ba17c","tarball":"https://registry.npmjs.org/@aiedx/firstbreakai/-/firstbreakai-0.3.0.tgz","fileCount":19,"unpackedSize":44563,"signatures":[{"keyid":"SHA256:DhQ8wR5APBvFHLF/+Tc+AYvPOdTpcIDqOhxsBHRwC7U","sig":"MEUCIA3wgHbLkpGehL4n+/lGcJX1e3n4/0fE9DnJmAnsvumlAiEAz/YjaCvFpQbA+1ARL41n2oM8iay7k4o72E1SkC2R5tM="}]},"_npmUser":{"name":"adssidhu86","email":"aiedx.org@gmail.com"},"directories":{},"maintainers":[{"name":"adssidhu86","email":"aiedx.org@gmail.com"}],"_npmOperationalInternal":{"host":"s3://npm-registry-packages-npm-production","tmp":"tmp/firstbreakai_0.3.0_1781504991945_0.6903688132209505"},"_hasShrinkwrap":false}},"time":{"created":"2026-06-03T15:54:48.740Z","modified":"2026-06-15T06:29:52.224Z","0.1.0":"2026-06-03T15:54:49.192Z","0.2.0":"2026-06-03T16:23:08.791Z","0.3.0":"2026-06-15T06:29:52.101Z"},"bugs":{"url":"https://github.com/thefirehacker/firstbreakai/issues"},"author":{"name":"First Break AI","email":"contact@bubblspace.com","url":"https://cohort.bubblnet.com"},"license":"MIT","homepage":"https://cohort.bubblnet.com","keywords":["ai","cohort","llm","machine-learning","huggingface","quarto","first-break-ai","mcp","inference","training","agentic-ai","ai-agents","ai-agent","agentic","model-context-protocol","mcp-server","claude","cursor","codex","ai-tutor"],"repository":{"type":"git","url":"git+https://github.com/thefirehacker/firstbreakai.git"},"description":"CLI + local MCP server for First Break AI — the first fully agentic AI cohort, where AI agents check your environment, validate your code, track progress, and answer questions. Powered by FetchLens.ai.","maintainers":[{"name":"adssidhu86","email":"aiedx.org@gmail.com"}],"readme":"# @aiedx/firstbreakai\n\nThe official CLI and local MCP server for [First Break AI](https://cohort.bubblnet.com) — **the first fully agentic AI cohort**.\n\nFirst Break AI is a cohort **run by AI agents**. AI agents answer your questions, validate your code against rubrics, and track your progress across the website, your terminal, and your IDE. The agentic backbone is powered by [FetchLens.ai](https://fetchlens.ai).\n\nThis package is how you put those **AI agents in your own terminal and IDE**: check your dev environment, track progress, validate exercises, scaffold a blog, and get AI-agent help — from the command line or any MCP-capable AI assistant (Cursor, Claude, Codex).\n\n## Install\n\n```bash\nnpm install -g @aiedx/firstbreakai\n```\n\nOr run without installing:\n\n```bash\nnpx @aiedx/firstbreakai doctor\n```\n\n**Requires:** Node.js >= 18\n\n## How AI agents run the cohort\n\nFirst Break AI is **completely agentic** — every part of the cohort is powered by AI agents, not static videos and quizzes:\n\n- **AI agents answer your questions** — the on-site assistant and `firstbreakai ask` know the syllabus, lessons, and your current step.\n- **AI agents validate your code** — `firstbreakai validate` runs deterministic checks against the rubric and reports exactly what passed.\n- **AI agents track your progress** — log in with Discord and your progress syncs across the CLI, the website, and the MCP server.\n- **AI agents live in your IDE** — run `firstbreakai mcp` (or use the remote MCP server) so Cursor, Claude Desktop, Claude Code, and OpenAI Codex can call cohort tools directly.\n- **Powered by [FetchLens.ai](https://fetchlens.ai)** — the agentic AI platform that provides the MCP server backbone, the AI-agent assistant, and authenticated progress tracking.\n\n## Commands\n\n| Command | Description |\n|---------|-------------|\n| `firstbreakai doctor` | Check your dev environment (Git, Python, Quarto, HF CLI, Node.js, Cursor) |\n| `firstbreakai status` | Show your cohort progress |\n| `firstbreakai done <step>` | Mark a step as complete |\n| `firstbreakai init` | Scaffold a new Quarto blog |\n| `firstbreakai open <page>` | Open a cohort page in your browser |\n| `firstbreakai next` | Open the next incomplete step |\n| `firstbreakai ask \"question\"` | Ask the FBA AI assistant |\n| `firstbreakai login` | Authenticate via Discord |\n| `firstbreakai validate <step>` | Run local validation checks for a step |\n| `firstbreakai mcp` | Start the MCP server for AI IDE integration |\n\n## Quick start\n\n```bash\n# 1. Check your environment\nfirstbreakai doctor\n\n# 2. Scaffold a blog\nmkdir my-ai-blog && cd my-ai-blog\nfirstbreakai init\n\n# 3. Preview it\nquarto preview\n\n# 4. Mark step 1 as done\nfirstbreakai done 1\n\n# 5. See what's next\nfirstbreakai next\n```\n\n## MCP server — give your AI agent cohort superpowers\n\nRun First Break AI as a local [Model Context Protocol](https://modelcontextprotocol.io) server so the **AI agent inside your IDE** (Cursor, Claude Desktop, Claude Code, OpenAI Codex) can call cohort tools directly — no context switching, fully agentic.\n\nAdd to your `.cursor/mcp.json` or Claude Desktop config:\n\n```json\n{\n  \"mcpServers\": {\n    \"firstbreakai\": {\n      \"command\": \"npx\",\n      \"args\": [\"@aiedx/firstbreakai\", \"mcp\"]\n    }\n  }\n}\n```\n\nThis exposes these agentic cohort tools to your AI agent:\n\n- `cohort_doctor` — environment check\n- `cohort_status` — progress overview\n- `cohort_done` — mark step complete\n- `cohort_validate` — run validation checks\n- `cohort_open` — open cohort pages\n- `cohort_next` — open next step\n- `cohort_ask` — ask the FBA AI assistant\n\n## Validation\n\n`firstbreakai validate <step>` runs deterministic, local checks — no LLM, no network required.\n\n**Step 1** (Quarto blog): checks `_quarto.yml`, `.qmd` files, Git repo, GitHub remote.\n\n**Step 2** (Run a model locally): checks `run.c`/binary, `config.json`, tokenizer, model weights.\n\nIf you're logged in (`firstbreakai login`), results are also saved to your profile on the FBA server.\n\n## Links\n\n- [Cohort homepage](https://cohort.bubblnet.com)\n- [Roadmap](https://cohort.bubblnet.com/roadmap)\n- [Lessons](https://cohort.bubblnet.com/lessons/)\n- [Checklist](https://cohort.bubblnet.com/checklist)\n- [Discord](https://discord.gg/hRPese4H3F)\n- [GitHub](https://github.com/thefirehacker/firstbreakai)\n\n## License\n\nMIT\n","readmeFilename":"README.md"}