# EdgeSpark Documentation

> Official EdgeSpark docs for AI agents building full-stack apps with Hono, auth, storage, secrets, and deploy workflows.

## Docs

- [CLAUDE](https://docs.edgespark.dev/CLAUDE.md)
- [AGENTS.md reference for EdgeSpark](https://docs.edgespark.dev/agents/agents-md.md): How AI agents should read the AGENTS.md (or CLAUDE.md / GEMINI.md) file in EdgeSpark projects and follow its instructions.
- [Aider and EdgeSpark](https://docs.edgespark.dev/agents/aider.md): Set up Aider to work with EdgeSpark — install EdgeSpark skills and docs MCP, load AGENTS.md with --read, and build with full context.
- [Amp and EdgeSpark](https://docs.edgespark.dev/agents/amp.md): Set up Amp (by Sourcegraph) to work with EdgeSpark — install the CLI, generate AGENTS.md, and give Amp the context it needs to build on EdgeSpark.
- [Claude Code and EdgeSpark](https://docs.edgespark.dev/agents/claude-code.md): Set up Claude Code to work with EdgeSpark — install the EdgeSpark plugin, generate CLAUDE.md, and follow the public Claude workflow.
- [Cline and EdgeSpark](https://docs.edgespark.dev/agents/cline.md): Set up Cline (VS Code extension) to work with EdgeSpark — install EdgeSpark skills and docs MCP, copy AGENTS.md to .clinerules, and build with full context.
- [OpenAI Codex and EdgeSpark](https://docs.edgespark.dev/agents/codex.md): Set up OpenAI Codex to work with EdgeSpark — install the EdgeSpark Codex plugin, generate AGENTS.md, and build with the public Codex workflow.
- [Continue and EdgeSpark](https://docs.edgespark.dev/agents/continue.md): Set up Continue (VS Code and JetBrains extension) to work with EdgeSpark — install EdgeSpark skills and docs MCP, then configure AGENTS.md as context.
- [Cursor and EdgeSpark](https://docs.edgespark.dev/agents/cursor.md): Set up Cursor to work with EdgeSpark — install the EdgeSpark Cursor plugin, generate AGENTS.md, and use the dedicated Cursor workflow.
- [The EdgeSpark declarative workflow](https://docs.edgespark.dev/agents/declarative-workflow.md): How EdgeSpark agents should work with declarative repo-owned defs, generated SDK types, pulled system schema, and migrations.
- [The EdgeSpark deploy and test loop](https://docs.edgespark.dev/agents/deploy-and-test.md): EdgeSpark has no local dev server. Use dry runs, cautious deploys, and live log inspection while public staging support is still coming soon.
- [Devin and EdgeSpark](https://docs.edgespark.dev/agents/devin.md): Set up Devin (by Cognition AI) to work with EdgeSpark — install EdgeSpark skills and docs MCP, generate AGENTS.md, and run autonomously.
- [Gemini CLI and EdgeSpark](https://docs.edgespark.dev/agents/gemini-cli.md): Set up Gemini CLI to work with EdgeSpark — install the extension, generate GEMINI.md, and use pre-built tools for EdgeSpark workflows.
- [GitHub Copilot and EdgeSpark](https://docs.edgespark.dev/agents/github-copilot.md): Set up GitHub Copilot with EdgeSpark — install the EdgeSpark Copilot plugin, generate AGENTS.md, and follow the public Copilot workflow.
- [Handling EdgeSpark errors autonomously](https://docs.edgespark.dev/agents/handling-errors.md): How to read and fix type errors, migration issues, build failures, SQL validation rejections, and runtime errors in EdgeSpark.
- [The EdgeSpark harness model](https://docs.edgespark.dev/agents/harness.md): How EdgeSpark enforces security, correctness, and safety at the platform level — guardrails that make AI-generated code safe by default.
- [EdgeSpark docs for AI agents](https://docs.edgespark.dev/agents/index.md): Learn how AI agents should use EdgeSpark: harness guardrails, declarative workflows, deploy loops, error handling, and setup paths.
- [OpenCode and EdgeSpark](https://docs.edgespark.dev/agents/opencode.md): Set up OpenCode to work with EdgeSpark — install the CLI, generate AGENTS.md, and give OpenCode the context it needs to build on EdgeSpark.
- [All other AI agents and EdgeSpark](https://docs.edgespark.dev/agents/other-agents.md): Set up all other AI agents with EdgeSpark using the generic skills, docs MCP path, and AGENTS.md instructions file.
- [All supported AI agents for EdgeSpark](https://docs.edgespark.dev/agents/supported-agents.md): EdgeSpark supports all AI agents: first-class plugin paths for Claude Code, Gemini CLI, OpenAI Codex, Cursor, and GitHub Copilot, plus generic AGENTS.md setups.
- [Minimal human input in EdgeSpark](https://docs.edgespark.dev/agents/when-to-ask.md): Agents should finish the task without interruptions by default, and stop only for login, secret entry, or clearly destructive actions.
- [Windsurf and EdgeSpark](https://docs.edgespark.dev/agents/windsurf.md): Set up Windsurf (by Codeium) to work with EdgeSpark — install EdgeSpark skills and docs MCP, copy AGENTS.md to .windsurfrules, and build with full context.
- [EdgeSpark CLI commands reference](https://docs.edgespark.dev/cli/commands.md): Reference for the current EdgeSpark CLI, including init, deploy, pull, db, storage, secrets, vars, auth config, and log commands.
- [Install EdgeSpark CLI and scaffold a project](https://docs.edgespark.dev/cli/installation.md): Install the EdgeSpark CLI, sign in with edgespark login, and scaffold the current full-stack project layout for Cloudflare deployments.
- [The EdgeSpark client object](https://docs.edgespark.dev/concepts/client-object.md): How the EdgeSpark runtime SDK works: database, auth, storage, secrets, vars, and request context imports available inside your handlers.
- [The EdgeSpark deploy pipeline](https://docs.edgespark.dev/concepts/deploy-pipeline.md): What edgespark deploy does: schema sync, TypeScript type check, esbuild bundle, route analysis, upload, and live deployment to Cloudflare Workers.
- [EdgeSpark environments](https://docs.edgespark.dev/concepts/environments.md): Understand the current EdgeSpark environment model: new projects start with one default production environment, while staging is coming soon.
- [How EdgeSpark works](https://docs.edgespark.dev/concepts/how-edgespark-works.md): How EdgeSpark works: the platform manages deployments and generated types, while your Hono app runs globally on Cloudflare's edge.
- [Path-based auth in EdgeSpark](https://docs.edgespark.dev/concepts/path-based-auth.md): EdgeSpark enforces authentication through route prefixes: /api/* requires login, /api/public/* is optional, and /api/webhooks/* skips auth.
- [EdgeSpark platform security](https://docs.edgespark.dev/concepts/platform-security.md): Learn how EdgeSpark enforces auth, validates runtime SQL, and isolates storage access before your handler code runs in production.
- [Manage EdgeSpark auth configuration](https://docs.edgespark.dev/guides/auth-config.md): Configure EdgeSpark auth providers, session settings, sign-up policy, vars, and secret refs through configs/auth-config.yaml.
- [Build auth UI with @edgespark/web](https://docs.edgespark.dev/guides/auth-ui.md): Build login, signup, and session-aware frontend flows in EdgeSpark with the @edgespark/web browser SDK and managed auth UI.
- [Authenticate users in EdgeSpark](https://docs.edgespark.dev/guides/authentication.md): Access the current user with auth.user, protect API routes, handle optional auth in public routes, and verify webhook signatures in EdgeSpark.
- [Bind a custom domain to your EdgeSpark project](https://docs.edgespark.dev/guides/custom-domains.md): Add your own hostname with edgespark domain add, publish the printed DNS records, and run edgespark domain verify to wait for activation and TLS.
- [Use the EdgeSpark database](https://docs.edgespark.dev/guides/database.md): Define Drizzle schema in your EdgeSpark repo, generate migrations, and query your Cloudflare D1 database with the runtime SDK.
- [Deploy an EdgeSpark project](https://docs.edgespark.dev/guides/deployment.md): Deploy your EdgeSpark project with edgespark deploy, validate with dry runs, and understand the current single-environment model.
- [EdgeSpark development workflow](https://docs.edgespark.dev/guides/development-workflow.md): Follow the full EdgeSpark development workflow for schema, storage, auth config, vars, secrets, generated types, and current deploys.
- [Error handling in EdgeSpark](https://docs.edgespark.dev/guides/error-handling.md): Patterns for validation errors, database errors, auth failures, and consistent JSON responses in modern EdgeSpark server routes.
- [Develop EdgeSpark apps locally with edgespark dev](https://docs.edgespark.dev/guides/local-mode.md): Run the full EdgeSpark stack on your machine with edgespark dev — local D1, R2, auth, hot reload, seed data, and a dev-mode auto-verify endpoint that replaces real email.
- [View EdgeSpark logs](https://docs.edgespark.dev/guides/logs.md): Use edgespark log tail for live EdgeSpark observability, background log capture, and real-time debugging of deployed requests.
- [EdgeSpark project structure](https://docs.edgespark.dev/guides/project-structure.md): How an EdgeSpark project is organized: server and web apps, repo-authored defs, generated files, auth config, and what never to edit manually.
- [Manage EdgeSpark secrets](https://docs.edgespark.dev/guides/secrets.md): Register, list, and delete EdgeSpark secrets from the CLI, then read them safely at runtime without sending values through LLM APIs.
- [Add social OAuth login to your EdgeSpark app](https://docs.edgespark.dev/guides/social-login.md): Enable Google, GitHub, GitLab, and Discord OAuth login in EdgeSpark through configs/auth-config.yaml, vars, and secrets.
- [Start an EdgeSpark project from a template](https://docs.edgespark.dev/guides/start-with-template.md): Clone an EdgeSpark template into a new platform project with one edgespark init command and finish setup with a short, conditional next-steps list.
- [Use EdgeSpark storage](https://docs.edgespark.dev/guides/storage.md): Declare EdgeSpark storage buckets in your repo, apply them with the CLI, and upload or serve files from the runtime SDK.
- [Test EdgeSpark applications](https://docs.edgespark.dev/guides/testing.md): Test EdgeSpark applications with unit tests and deployed integration checks that fit the current production-first workflow.
- [Manage EdgeSpark variables](https://docs.edgespark.dev/guides/variables.md): Set EdgeSpark runtime vars for non-sensitive configuration, type the keys in runtime.ts, and read them safely in server code.
- [EdgeSpark docs for AI agents](https://docs.edgespark.dev/index.md): Learn how to build and deploy full-stack Hono apps on Cloudflare with EdgeSpark, using D1, R2, auth, secrets, and agent-ready CLI workflows.
- [EdgeSpark quickstart for full-stack Cloudflare apps](https://docs.edgespark.dev/quickstart.md): Create your first EdgeSpark full-stack app, install server and web dependencies, and deploy a Hono project to Cloudflare with D1, R2, and auth.
- [EdgeSpark platform limits](https://docs.edgespark.dev/reference/limits.md): Reference the current EdgeSpark platform limits across compute, database, storage, secrets, vars, request size, and deployment constraints.
- [auth — EdgeSpark authentication](https://docs.edgespark.dev/sdk/auth.md): Access the current authenticated user with the auth import from edgespark/http and understand auth behavior by route pattern.
- [db — EdgeSpark database](https://docs.edgespark.dev/sdk/database.md): Query your EdgeSpark D1 database with the runtime db import. Use Drizzle selects, inserts, updates, deletes, relations, and batch operations.
- [EdgeSpark SDK reference](https://docs.edgespark.dev/sdk/overview.md): Reference the EdgeSpark server runtime imports and the @edgespark/web browser SDK used by the current full-stack scaffold.
- [secret — EdgeSpark secrets](https://docs.edgespark.dev/sdk/secrets.md): Read encrypted EdgeSpark secrets at runtime after registering keys through the CLI and entering values through a secure browser flow.
- [storage — EdgeSpark file storage](https://docs.edgespark.dev/sdk/storage.md): Use the EdgeSpark storage import for R2 buckets, S3 URIs, metadata reads, pagination, and presigned upload or download URLs.
- [vars — EdgeSpark variables](https://docs.edgespark.dev/sdk/vars.md): Read plain EdgeSpark runtime vars with vars.get, keep allowed keys typed in runtime.ts, and separate vars from secrets safely.
- [web — EdgeSpark browser SDK](https://docs.edgespark.dev/sdk/web.md): Use the @edgespark/web browser SDK for same-origin auth, managed auth UI, session helpers, errors, locales, and API requests.
- [Build an EdgeSpark REST API with Hono and D1](https://docs.edgespark.dev/tutorials/build-a-rest-api.md): Build a production-style EdgeSpark REST API with Hono, D1, auth, presigned uploads, secrets, and a deployable server scaffold.

## Optional

- [GitHub](https://github.com/edgesparkhq)
- [Dashboard](https://edgespark.dev)