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 embedded systems engineers

OTA Update Safety Check Agent

Prove an update recovers safely from every failure mode before it reaches the field

Quality controlRisk managementIT and developmentProduct development
See how it worksGet access
OTA Update Safety Check Agent: what goes in, what the agent does and what you get

What it does

One bad over-the-air update can leave thousands of devices stuck. Before a release, this agent runs the update on a pool of test devices and breaks it on purpose. It cuts power at several points during download, writing and reboot. It sends corrupted and truncated images, an image signed with the wrong key and one for the wrong hardware version. It tries a rollback from the new version to the old. After each run it checks that the device boots to a working version and that stored data is intact, and it lists every failure. When a fix is made it reruns the whole matrix. The lead approves the rollout and each stage. Edge case: a device recovers only after a second reboot, so the agent counts it as a failure.

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, continueApprovedNo 1 STARTS WHEN Release candidate update package ready 2 USES A TOOL Prepare test devices on the previous version 3 USES A TOOL Run a normal update and check boot and data 4 USES A TOOL Cut power at set points during download, write andreboot 5 USES A TOOL Send corrupt, truncated, wrong-key andwrong-hardware images 6 USES A TOOL Test rollback to the previous version 7 DOES Record boot result, data check and recovery time foreach case 8 CHECKS THE RESULT Did every case recover to a working version withouthelp? If not: list failures for the developer, wait for a fix,and rerun the entire matrix. Back to step 4. 9 YOU APPROVE Lead approves the staged rollout plan 10 RESULT Safety report with all cases and the rollout plan
Read the steps as a list
  1. Release candidate update package ready
  2. Prepare test devices on the previous version
  3. Run a normal update and check boot and data
  4. Cut power at set points during download, write and reboot
  5. Send corrupt, truncated, wrong-key and wrong-hardware images
  6. Test rollback to the previous version
  7. Record boot result, data check and recovery time for each case
  8. Did every case recover to a working version without help?If not: list failures for the developer, wait for a fix, and rerun the entire matrix. Back to step 4.
  9. Lead approves the staged rollout planThe agent waits here for your OK.
  10. Safety report with all cases and the rollout plan

How it decides

It marks a case as failed unless the device returns to a working version on its own with data intact within the allowed time.

  • Count a case as failed if manual intervention is needed
  • Count a case as failed if recovery takes over 3 reboots
  • Require every power-cut point to be tested at least 3 times
  • Block release if stored settings change after a failed update

Make it yours

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

  • Failure cases in the matrix
  • Number of test devices (default 6)
  • Repeats per power-cut point (default 3)
  • Maximum reboots allowed (default 3)
  • Rollout stage sizes

What keeps you in control

It always asks you first

  • Lead approves the rollout and each stage
  • Lead approves any exception for a known failure

Hard limits

  • Never push test images to field devices
  • Never skip the rollback test

It stops when

  • Done: all cases pass and the rollout plan is approved
  • Stop: a case bricks a device and no fix is ready

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 agent ran 148 cases on 6 devices. Two failed. A power cut at 94% of the flash write left a device in a boot loop, and a wrong-key image was accepted on one board. The developer fixed both. The rerun of all 148 cases passed except one slow recovery of 4 reboots, so the check failed. After a boot counter fix the third run passed. The lead approved a 1%, 10%, 100% rollout.

More agents for embedded systems engineers

Quality control · IT and development

Firmware Release Candidate Regression Agent

Show every regression in a release candidate and trace each to its cause

11 steps2 checks6 jobs
Quality control · IT and development

Hardware-in-the-Loop Test Run Agent

Run the full rig suite unattended and classify each failure with evidence

10 steps2 checks8 jobs
Analysis · Monitoring

Memory and Stack Budget Agent

Keep flash, RAM and stack use inside budget with a clear cause for each increase

10 steps2 checks2 jobs
Analysis · IT and development

RTOS Timing and Priority Review Agent

Remove missed deadlines and priority inversions with measured proof

10 steps2 checks5 jobs