Complete AI Training
Sign inGet my AI kit

Your job's AI kit

Get your AI kit

Tell us who you are and what you do. We show you your kit right away and email you the link: skills, prompts, AI agents, MCP servers and courses for your job.

500+ jobs ready, and we make a kit for any other job. No payment needed to look.

Share

AI agent for backend developers

Feature Flag Cleanup Agent

Remove stale feature flags and their dead code with a passing test suite and a reviewed pull request.

CodingIT and development
See how it worksGet access
Feature Flag Cleanup Agent: what goes in, what the agent does and what you get

What it does

Feature flags are meant to be temporary, but old ones pile up, hiding dead code and making behavior hard to reason about. The agent scans the flag system for flags that have been fully on or fully off for a set time, finds every code reference, and drafts the change that removes the flag and keeps the winning path. It runs the test suite on the branch. If tests fail, it reads the failure, revises the change and runs the tests again. When the tests pass, it opens a pull request with a summary of what was removed and why. Edge case: a flag is fully on but still read in a rarely used admin path, so the agent keeps that branch and notes it.

How it works

Follow the arrows from top to bottom. The orange dashed arrow is the loop: when a check fails, the agent goes back and tries again.

Start and resultWhat it doesA check on its own workWaits for your OKGoes back and retries
Yes, continueYes, continueApprovedNoNo 1 STARTS WHEN Weekly flag scan 2 USES A TOOL List flags and how long each has been fully on oroff 3 DOES Select eligible flags with no recent change 4 USES A TOOL Search the code for every reference 5 DOES Draft the change that removes the flag and keeps thelive path 6 USES A TOOL Run the test suite on the branch 7 CHECKS THE RESULT Do all tests pass? If not: Read the failure, revise the change and rerun.Back to step 5. 8 DOES Check the diff for references in config, docs andadmin paths 9 CHECKS THE RESULT Is every reference handled? If not: Update the missing references and rerun tests.Back to step 5. 10 YOU APPROVE Developer approves the pull request 11 RESULT Merged cleanup and updated flag list
Read the steps as a list
  1. Weekly flag scan
  2. List flags and how long each has been fully on or off
  3. Select eligible flags with no recent change
  4. Search the code for every reference
  5. Draft the change that removes the flag and keeps the live path
  6. Run the test suite on the branch
  7. Do all tests pass?If not: Read the failure, revise the change and rerun. Back to step 5.
  8. Check the diff for references in config, docs and admin paths
  9. Is every reference handled?If not: Update the missing references and rerun tests. Back to step 5.
  10. Developer approves the pull requestThe agent waits here for your OK.
  11. Merged cleanup and updated flag list

How it decides

A flag is eligible when it has been at 100 percent on or off for 30 days and has a known owner. The agent keeps the code path matching the current state and removes the other.

  • Flags at 100 percent for 30 days or more are eligible
  • A flag with active experiments or a changing rollout is skipped
  • Remove at most 5 flags per pull request
  • If tests fail three times, stop and hand the flag to the developer

Make it yours

Every agent is a starting point. You choose these settings for your own situation.

  • Days at full on or off before eligible (default 30)
  • Maximum flags per pull request (default 5)
  • Repositories in scope
  • Flags protected from cleanup

What keeps you in control

It always asks you first

  • Developer approves and merges the pull request

Hard limits

  • Never merges or deploys code
  • Never touches a flag with a changing rollout

It stops when

  • Done: pull request approved and merged
  • Stop: tests keep failing after three attempts, so the developer takes over

Set it up

We guide you through the set-up, step by step

Members get the full set-up guide for this agent. No technical skills needed: you copy, paste and upload.

10 minto set it up in your AI
5 AIsChatGPT, Claude, Copilot, Gemini, Grok
  • One set of instructions to paste into your AI, with the clicks for ChatGPT, Claude, Microsoft 365 Copilot, Gemini and Grok
  • The agent then walks you through connecting your own data, one source at a time
  • A downloadable copy with the flow chart, the rules and the full guide
Get access to this agent

An example run

What happensThe scan found 8 flags on for over 60 days. For the new checkout flag, the agent removed 14 references, but 2 tests failed because a mock still set the flag. It revised the test setup and reran, and all 312 tests passed. A final check found one reference in an admin script, which it handled. The developer approved the pull request.

More agents for backend developers

Coding · Quality control

API Contract Drift Repair Agent

Schema, clients and implementation agree, without surprise breaking changes.

8 steps1 check3 jobs
Quality control · Coding

Database Migration Rehearsal Agent

Migrations rehearsed safely with schema and data intact.

7 steps1 check4 jobs
Coding · Quality control

API Pagination Recovery Agent

Syncs that return every record exactly once.

8 steps1 check3 jobs
Quality control · Product development

Product Supportability Review Agent

Every feature can be diagnosed by support when it fails.

9 steps1 check1 job