{"slug": "i-turned-my-claude-md-files-into-a-marketplace-and-an-api-your-agent-can-use", "title": "I turned my CLAUDE.md files into a marketplace (and an API your agent can use)", "summary": "A developer launched Makimono, a marketplace for CLAUDE.md instruction files that help AI coding agents skip trial-and-error by reusing proven project playbooks. The platform features an open, no-auth API for agents to search and retrieve files autonomously, with a token-savings metric to guide decisions. Listings are submitted via a single prompt to an agent, and the project aims to build a flywheel of shared developer judgment.", "body_md": "Every new project, my coding agent burns tokens re-deriving the same setup: the tech\n\nchoices, the gotchas, the \"actually do it this way\" rules I already figured out last time.\n\nSo I built **Makimono** — a marketplace for the finished instruction files (CLAUDE.md,\n\nskills, dev playbooks) that make an agent skip the trial-and-error and go straight to done.\n\n🔗 [https://makimono-md.vercel.app/en](https://makimono-md.vercel.app/en)\n\nBuilding something from scratch with an AI agent costs tokens — and tokens cost money.\n\nA lot of that cost is the agent figuring out *how* to build the thing: which libraries,\n\nwhich gotchas to avoid, which conventions to follow. If someone already solved that and\n\nwrote it down as a CLAUDE.md, feeding that file to your agent skips the figuring-out phase\n\nentirely.\n\nRough shape of the trade:\n\nYou pay for the answer, not the search for the answer.\n\n**1. An open, no-auth API.**\n\nNo API key required, CORS enabled, and a `/llms.txt`\n\nat the root describing how to use it.\n\nThe intent is that your coding agent can search Makimono and pull the right playbook by\n\nitself — no human has to browse a marketplace mid-task.\n\n```\n# Find a scroll for what you're building\ncurl \"https://makimono-md.vercel.app/api/v1/search?q=discord+bot\"\n\n# Read the top result, then load its raw body\ncurl \"https://makimono-md.vercel.app/api/v1/files/{slug}/raw\"\n```\n\nEach result includes an `roi`\n\nfield (tokens saved per token spent reading the file) so an\n\nagent can decide for itself whether reading a given scroll is worth it before committing to\n\nit.\n\n**2. Listing your own file is one pasted prompt.**\n\nInstead of a form, you paste a prompt into your own agent (Claude Code, etc.). It:\n\n~1 minute, and nothing goes out without you explicitly approving each item.\n\n→ [https://makimono-md.vercel.app/en/contribute](https://makimono-md.vercel.app/en/contribute)\n\nThere's also a Claude Code plugin that wires \"find a scroll before building\" and \"publish\n\nmine when I'm done\" in as skills:\n\n[https://github.com/seisaku-team-org/makimono-plugin](https://github.com/seisaku-team-org/makimono-plugin)\n\nA \"skill\" (SKILL.md) is a narrow, packaged capability — it teaches an agent how to do one\n\nspecific thing, in a fixed format. A CLAUDE.md-style playbook is broader: it's the\n\naccumulated judgment behind an entire project — the tech choices, the gotchas, the\n\n\"actually, do it this way\" calls that only show up after you've been burned once. It's\n\nplain Markdown, not a packaging spec, so it works with any agent that can read text.\n\nSkills are welcome on Makimono too (some listings are skills), but what the marketplace is\n\nreally built to collect is judgment, not just capability — which is also why the\n\ntoken-savings metric (scratch vs. with-scroll) makes sense here in a way it barely does for\n\na single narrow skill.\n\nListings and usage reinforce each other:\n\n`/api/v1/report`\n\nendpoint)Each turn compounds. A scroll collapses one agent's trial-and-error *loop* into a single\n\npass; multiply that across contributors submitting proven playbooks and it becomes a\n\nflywheel for the whole catalog.\n\nThis just launched (from Japan — a lot of current listings are in Japanese, since that's\n\nwhere the first batch of contributors are). Listing count is small and I'm not going to\n\npretend otherwise. What I'm most interested in is whether the \"agents transact\n\nautonomously via an open API\" design holds up as more scrolls get added — and whether\n\nEnglish CLAUDE.md files show up to make the catalog useful for non-Japanese-speaking\n\nagents too.\n\nIf you've got a CLAUDE.md that reliably makes an agent do the right thing, I'd genuinely\n\nlike to see it on the shelf: [https://makimono-md.vercel.app/en/contribute](https://makimono-md.vercel.app/en/contribute)\n\nFeedback on the API shape, the listing flow, or anything else is very welcome.", "url": "https://wpnews.pro/news/i-turned-my-claude-md-files-into-a-marketplace-and-an-api-your-agent-can-use", "canonical_source": "https://dev.to/_87fba493a84f2b519cd0/i-turned-my-claudemd-files-into-a-marketplace-and-an-api-your-agent-can-use-50me", "published_at": "2026-07-21 22:26:40+00:00", "updated_at": "2026-07-21 22:29:22.861210+00:00", "lang": "en", "topics": ["ai-agents", "developer-tools", "artificial-intelligence", "ai-products"], "entities": ["Makimono", "Claude Code", "seisaku-team-org"], "alternates": {"html": "https://wpnews.pro/news/i-turned-my-claude-md-files-into-a-marketplace-and-an-api-your-agent-can-use", "markdown": "https://wpnews.pro/news/i-turned-my-claude-md-files-into-a-marketplace-and-an-api-your-agent-can-use.md", "text": "https://wpnews.pro/news/i-turned-my-claude-md-files-into-a-marketplace-and-an-api-your-agent-can-use.txt", "jsonld": "https://wpnews.pro/news/i-turned-my-claude-md-files-into-a-marketplace-and-an-api-your-agent-can-use.jsonld"}}