Plan-First Feature Implementation

Make an AI coding agent explore and plan before touching code: file-level plan, risks, and edge cases; implementation gated on your approval.

I want to implement this feature: [DESCRIBE THE FEATURE: the user-visible outcome, not the implementation]

Constraints: [e.g. "no new dependencies", "must work with our existing auth", "ship behind a feature flag"]

Work in two phases. Do NOT write or edit any code in phase 1.

**Phase 1: Explore and plan:**
1. Read the relevant parts of the codebase first. List the files you examined and what each contributes to this feature.
2. Produce the plan:
   - Approach in 3-5 sentences, including the one key design decision and why.
   - File-by-file change list: file path → what changes and why. Mark new files as NEW.
   - Data/schema/API contract changes, if any: exact shapes.
   - Edge cases this must handle: empty states, concurrent use, failures of any external call, permissions.
   - What could break: existing behavior this touches, and how we'll know it didn't (which tests cover it, or which need writing).
   - Out of scope: what I might expect that you're deliberately NOT doing, so we agree now.
3. If the feature description is ambiguous in a way that changes the design, ask me the specific question instead of picking silently.

Stop after the plan and wait for my approval.

**Phase 2: after I approve:** implement exactly the approved plan. If reality diverges from the plan mid-implementation (an assumption was wrong), stop and tell me what changed rather than improvising a different design. Finish by running the tests and reporting results honestly.

How to use

This is the explore-plan-code workflow that agent vendors themselves recommend: the gate between planning and implementation is where you catch wrong assumptions cheaply. Works in Claude Code, Cursor, and any agentic tool; for chat-only use, paste the relevant files where it says to read the codebase.

More coding prompts

Write the commit message for this change.

Diff:
"""
[PASTE THE STAGED DIFF]
"""

Why I made this change: [THE REASON, THE TICKET, THE BUG REPORT, or "you infer it"]
Convention: [Conventional Commits / this repo's existing style, pasted below / plain]
Recent commits from this repo, to match style:
"""
[PASTE 5-10 RECENT COMMIT SUBJECT LIN

Commit Message

Write a commit message that explains why the change was made, in Conventional Commits format, split into separate commits when needed.

Codingbeginner

Help me recover from a git mistake without making it worse.

What I was trying to do: [THE GOAL]
What I ran: [THE EXACT COMMANDS, IN ORDER]
What happened instead: [THE OUTPUT OR THE STATE NOW]
Has this been pushed or shared: [YES/NO, and to which branch and whether anyone else has pulled]
Uncommitted work I cannot lose: [WHAT AND WHERE, o

Undo a Git Mistake

Recover from a bad commit, force push, wrong branch, or lost work with a reversible plan and the exact commands, explained before you run them.

Codingintermediate

Handler code, routes, and models:
"""
[PASTE THE ROUTE DEFINITIONS, HANDLERS, REQUEST AND RESPONSE TYPES, VALIDATION SCHEMAS, AND MIDDLEWARE]
"""

Generate an OpenAPI 3.1 specification from the code above.

API name, version, and base URL: [DETAILS]
Auth scheme: [BEARER JWT / API KEY / OAUTH / SESSION COOKIE, and where it is enforced]
Con

OpenAPI Spec From Code

Generate an accurate OpenAPI 3.1 spec from handler code, including error responses and auth, with gaps flagged instead of invented.