Plan With the Agent First, Then Verify Its Work Yourself

Melvin Vivas · X post · 2026-08-21 · Open on X

Topics: AI Dev Tools & Productivity, AI Agents, Tool Use & MCP · Level: beginner

Summary

The creator ran Cursor's agent for a few hours from the iPhone app. Before it started building, he worked through the plan with the agent to make sure every spec from the product owner was covered. The last step, verifying the result, still had to be done by hand.

Key points

• Work through the implementation plan with the agent before it starts writing code.
• Check that the plan covers every spec from the product owner or stakeholder before you let the agent run.
• Long agent runs can be started and left running from Cursor's iPhone app.
• Final verification is still a human job. Set aside time to test the result by hand.

Resources mentioned

• Cursor · tool · x.com · free · recommended by both Bashiri Smith & Melvin Vivas
  AI-native code editor (VS Code with AI built in).
  Also in: GLM 5.3 and GLM 5.3 Flash now in Cursor (Melvin Vivas on X · notes), Grok Bot Can Now Hand Off Coding Tasks to Cursor (Melvin Vivas on X · notes), Why Plan Mode Still Matters in AI Coding Assistants (Codex, Claude, Cursor) (Melvin Vivas on X · notes), Why Plan Mode Still Matters in AI Coding Agents (Codex, Claude, Cursor) (Melvin Vivas on X · notes) and 121 more

Try this

• Before the agent builds anything, review its plan against the product specs.
• Test the agent's output by hand before you call the work done.