Add Cursor Cloud specific instructions to AGENTS.md

[olliethedev]

Summary

• Adds ## Cursor Cloud specific instructions section to AGENTS.md with quick-reference notes, gotchas, and service startup guidance for future cloud agents.

Type of change

• Chore / refactor / tooling

Checklist

• pnpm build passes
• pnpm typecheck passes
• pnpm lint passes
• Tests added or updated (unit and/or E2E) — N/A, no code changes
• Docs updated (docs/content/docs/) if consumer-facing types or behavior changed — N/A
• All three codegen-projects create successfully and pass E2E tests — N/A
• New plugin: submission checklist in CONTRIBUTING.md completed — N/A

Screenshots

Docs site landing page
Blog plugin docs page
btst_docs_site_demo.mp4

_{To show artifacts inline, enable in settings.}

  

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
better-stack-docs	Ready	Preview, Comment	Apr 9, 2026 11:01pm
better-stack-playground	Ready	Preview, Comment	Apr 9, 2026 11:01pm

[cursor]

Security Assessment — No High-Confidence Vulnerabilities Found

This PR adds 25 lines of developer workflow documentation to AGENTS.md. There are no code changes, no new dependencies, and no logic modifications. The full threat checklist was evaluated:

Injection (SQL, command, template, path traversal): None. All additions are static prose with no interpolated values or shell constructs.

Authn/authz bypasses: None. No authentication or authorization code is touched.

Secrets handling / token leakage: The new text mentions that OPENAI_API_KEY belongs in playground/.env.local. No actual secret value is committed; the guidance follows standard Next.js convention where .env.local is gitignored by default. Low-risk documentation note only.

Unsafe deserialization, SSRF, XSS, request forgery: None. No runtime code paths are introduced or modified.

Dependency / supply-chain risk: No new dependencies are added. The referenced scripts/codegen/setup-nextjs.sh script (which runs pnpm dlx shadcn@latest) is a pre-existing file and is not modified by this PR.

Prompt / agent-instruction injection: The additions are developer ergonomics notes (pnpm version pin, build ordering, local port numbers). No instructions are added that would cause an AI agent to exfiltrate data, bypass access controls, or take unintended actions.

No findings require remediation. Safe to merge from a security perspective.

  

_{Sent by Cursor Automation: Find vulnerabilities}