TPEE Toptronic® Prompt Engineering Editor

7. IDEs and Coding Agents

Use Cursor, VS Code, Claude Code, Codex CLI, and other tools with the right prompt shape.

By Jacques Botte, founder of Toptronic®. Last updated 19 September 2026.

The lesson

Each IDE or agent reads instructions differently. TPEE's IDE selector formats the same intent for Cursor, Claude Code, ChatGPT, Copilot, Windsurf, and others.

For coding agents, specify the working directory, files to inspect, tests to run, and what must not be modified.

Remember: TPEE formats and stores the prompt. The target tool performs the coding only after you paste or export the prompt into that tool.

A platform engineer formats a coding-agent prompt in TPEE, naming the working directory, the service files to inspect, the test command to run and the deployment scripts that must not change.

An automation engineer specifies the motion-control folder, the test suite and the safety modules to leave untouched before exporting the prompt into a coding agent.

A rail signalling developer writes a prompt that names the interlocking source files, the required unit tests and an explicit list of modules that must not be modified.

A gameplay programmer shapes an agent prompt for a collision bug, listing the directory, the reproduction steps, the tests to run and the rendering code that must stay untouched.

A backend engineer at a payments company drafts a coding-agent prompt that fixes one failing test, names the module scope and forbids changes to authentication code.

A clinical systems developer prompts a coding agent to add a validation rule, stating the working directory, the interface tests and the audit-logging module that must not be altered.

A school IT administrator writes an agent prompt for a timetable script, naming the file scope, the checks to run and the unrelated booking modules to leave alone.

A controls developer at a factory drafts an agent prompt for a logic export fix, listing the project folder, the simulation test and the safety routines that must not change.

A security tooling developer formats a coding-agent prompt that adds a log-parsing function, naming the files to inspect, the tests to run and the credential-handling code that stays untouched.

A junior firmware developer uses TPEE to write an agent prompt for a sensor driver fix, including the build target, the test harness and the communication module that is off limits.

Check yourself

Question 1: Why does TPEE have an IDE selector?
  1. To connect MCP servers
  2. Different AI tools prefer different prompt formats — correct
  3. To compile Rust code
  4. To start Cursor automatically

Answer: Different AI tools prefer different prompt formats

Formatting should match the target tool the user will paste into.

Question 2: What should a coding-agent prompt include?
  1. Working directory, files, tests, and what must not change — correct
  2. Only a vague command
  3. Secret keys
  4. No acceptance criteria

Answer: Working directory, files, tests, and what must not change

Agentic coding needs clear scope and verification instructions.

Question 3: What remains true about TPEE and coding agents?
  1. TPEE runs every coding agent
  2. TPEE commits code automatically
  3. TPEE opens pull requests by itself
  4. TPEE formats prompts; the separate tool performs work after manual handoff — correct

Answer: TPEE formats prompts; the separate tool performs work after manual handoff

TPEE remains a local prompt editor and training tool.

← Previous lesson · All 91 lessons · Next lesson →

The full course — 91 lessons and 273 quiz questions — ships inside the app. Get TPEE to study it offline.