# SpecRouter — full content ## Site metadata - Site name: SpecRouter - URL: https://specrouter.dev - Tagline: Ship AI agents you actually own. - Short description: SpecRouter generates a complete Spec-Driven Development flow for OpenCode or Claude Code — agents, skills, project context. Plain markdown. No opinionated defaults. MIT-licensed open source. - License: MIT - Source: https://github.com/specrouter - Contact: hi@specrouter.dev - Pricing page: https://specrouter.dev/pricing ## Hero Section label: Customizable SDD flow generator. Headline: Ship AI agents you actually own. Body: SpecRouter generates a complete Spec-Driven Development flow for OpenCode and Claude Code — agents, skills, project context. Plain markdown. No opinions baked in. You write the prompts, you ship the ZIP, you keep the code. Primary CTA: Generate my ZIP → /wizard Secondary CTA: See pricing → #pricing Promises (under the hero): - No signup - Bring your own API key - Free during beta - MIT licensed ZIP preview shows: AGENTS.md, .opencode/agents/specrouter-dev.md (orchestrator), .opencode/skills/bootstrap/SKILL.md, .opencode/agents/phase-hu.md, .opencode/agents/phase-spec.md, .opencode/agents/phase-code.md, .opencode/agents/phase-qa.md, docs/project-context.md, .env.example, README.md. Example prompt shown in the hero (orchestrator): # SpecRouter.dev (Orchestrator) ## When you act After `/bootstrap` (Claude Code) or `@bootstrap` (OpenCode) hands off. ## What you produce Four artifacts across four phases: - Phase 1 (HU): `docs/historias/HU--.md` - Phase 2 (SPEC): `docs/specs/SPEC-.md` - Phase 3 (CODE): the implementation - Phase 4 (QA): a QA report with PASS/WARN/FAIL verdict ## How you work 1. Inspect the project on disk to determine the current phase. 2. Invoke the phase sub-agent (phase-hu, phase-spec, phase-code, phase-qa). 3. Pause for human approval between phases. ## Built for Generates ready-to-use ZIPs for: OpenCode (available today), Claude Code (available today). On the roadmap: Cursor, Codex, Kiro, Continue.dev. ## How it works From blank repo to your AI team in 5 minutes. Three steps. No GitHub login. No account. No telemetry. Step 01 — Pick your tool & flow. OpenCode or Claude Code. Choose strict gates between phases, or relaxed ones. The wizard remembers your choices — re-run anytime to iterate. Step 02 — Customize every prompt. Describe each agent in one sentence. The LLM drafts the system prompt — you edit, accept, or rewrite. Or skip AI and write them yourself. You stay in control. Step 03 — Download, install, ship. One ZIP with all your customized agents, skills, and project files. Drop it in your repo, add your API key, run @bootstrap. Five minutes from zero to working SDD. ## The positioning A customizable alternative to opinionated SDD frameworks. Most agent configuration tools ship a curated preset and ask you to live with it. SpecRouter ships a generator and a blank canvas. Point 1 — You define the agents. SpecRouter ships a generator, not a curated preset. Every agent prompt starts blank — or from a starter you can rename, rewrite, or delete. No persona baked in. Point 2 — You own the prompts. The ZIP is plain markdown. Edit it, fork it, version it in your repo. SpecRouter does not phone home, inject telemetry, or wrap your agents in a proxy. MIT-licensed code, all of it. Point 3 — You compose the flow. Pick strict gates or relaxed ones. Bundle the optional local dashboard. Add or remove phase sub-agents as your project grows. SpecRouter fits your team, not the other way around. ## Who it's for Built for the people who refuse to rent their tools. Three audiences. One generator. Zero opinions baked into the output. Solo dev — Ship with structure, not chaos. Tired of copy-pasting prompt fragments between projects. Generate a clean SDD flow once, version it in your repo, and re-import it next time. Team lead — Standardize without dictating. Every teammate ships a slightly different AI flow. SpecRouter gives you a baseline the team agrees on, with room for each project to diverge. Enterprise / OSS maintainer — Open prompts, open code. Auditable. Forkable. No proxy. SpecRouter generates artifacts you can review line by line and ship to your compliance team without translation. ## Dashboard (optional) See your SDD flow at a glance. Add the SpecRouter dashboard to your ZIP and get a local web UI for your project's SDD state — kanban of HUs, recent activity, full-text search across docs/, and detail views per HU, spec, and QA report. Runs entirely on your machine. Highlights: Local (localhost:7777, no cloud, no account, no telemetry), Zero infra (one Node.js process, one SQLite file), Auto-started (OpenCode / Claude Code spawns the server), Opt-in (off by default). ## Pricing Free today. Fair when we ship Pro. The generator stays free. Pro funds the hosted features so the open-source core keeps moving. Free ($0) — Forever. The full generator. All four phases. Both tools. Use it as long as you want. Unlimited ZIP downloads, OpenCode + Claude Code ZIPs, AI-assisted prompt customization (BYO key), Local dashboard (opt-in), MIT-licensed source on GitHub. Pro (TBA) — Coming soon. For teams that want hosted prompt libraries, shared dashboards, and priority AI refinement. Everything in Free, hosted team prompt library, shared dashboard across teammates, faster AI refinement queue, priority support on GitHub issues. No pricing set yet — we will email you when it ships. ## Built different We don't do the usual SaaS things. Most agent tools hide their hooks behind a marketing page. We'd rather show our work. - Open source, MIT. Source on GitHub. License is MIT. No proprietary hooks, no SaaS-only features, no upgrade wall for the things the generator already does. - No telemetry. We do not track what you generate, what you download, or what you ship. The wizard runs in your browser; the LLM customization step uses your key in memory and discards it. - No vendor lock-in. The ZIP is plain markdown. Edit it in your repo. Fork it. Ship it without us. We win by being useful, not by holding your agents hostage. - Bring your own key. We never see, store, or proxy your LLM API key. Pay your model provider directly — pay cents, not dollars. No middleman markup. ## FAQ Q: What tools does SpecRouter support? A: OpenCode and Claude Code today. Both consume plain markdown for agents and skills, so the generated ZIP drops in without extra adapters. Cursor, Codex, Kiro, and Continue.dev are on the roadmap — vote on the next one by opening a GitHub issue. Q: Do I need an account or a subscription? A: No. SpecRouter is a generator that runs in your browser — no login, no signup, no server-side library. You bring your own LLM API key for the prompt-customization step; the key is used in memory and immediately discarded. The generated ZIP is yours. Q: How is SpecRouter different from Cursor, Kiro, or other agent-configuration tools? A: Most agent-configuration tools ship a curated preset — their authors' opinions baked in — and ask you to live with it. SpecRouter ships a generator and a blank canvas: every agent prompt is yours to define, rename, or delete. No preset persona, no routing rules you didn't pick, no telemetry wrapper around your agents, no SaaS lock-in. Q: Can I edit the generated agents after downloading? A: Yes — that's the whole point. The ZIP contains plain markdown files. Edit them in your repo, commit them, ship them. The next time you run SpecRouter you can re-import your edited prompts or start fresh. Q: Will the generator stay free? A: Yes. The generator is MIT-licensed open source and will stay free forever. A Pro tier is on the roadmap for hosted team features (shared prompt library, hosted dashboards, priority AI refinement), but the generator, the ZIP output, and the core flow remain free. Q: What does Pro cost? A: We have not set Pro pricing yet — we will email waitlist subscribers when we do. Pro is opt-in only; existing free users keep everything they have today even when Pro ships. Q: Is my API key safe? A: Yes. Your key is held in your browser's memory only, used for the single customization request you triggered, and never written to disk, sent to our servers, or included in the ZIP. Three independent regression suites verify this — see the security tests in the repo. Q: Does SpecRouter work offline? A: The generator runs entirely in your browser and works without network access for the UI. The only network call is the optional AI customization step (POST /api/generate-agent), which requires your key and a reachable model provider. ## Final CTA Your next SDD ZIP is five minutes away. Pick your tool. Pick your flow. Generate. You keep the prompts, you keep the ZIP, you keep the code. Primary CTA: Generate my ZIP → /wizard Secondary CTA: Read the source → https://github.com/specrouter Trust line: Free during beta · MIT licensed · No signup ## Navigation Top nav: SpecRouter logo (home), "How it works" (anchor), "Pricing" (anchor), "FAQ" (anchor), language switcher, GitHub link, "Try it now" CTA. Footer columns: - Brand block: SpecRouter logo + tagline ("The customizable SDD flow generator for OpenCode and Claude Code"). - Product: Generate a ZIP, Pricing, How it works, Dashboard, FAQ. - Resources: GitHub, llms.txt, llms-full.txt, sitemap.xml, Contact. - Roadmap: Cursor support, Codex support, Hosted dashboards (Pro), Team prompt library. Copyright line: © SpecRouter. MIT licensed.