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 biomedical engineers

Device Software Anomaly Triage Agent

Every open anomaly has a severity, a hazard check and a reviewed release decision

Risk managementDecisionsScience and researchProduct development
See how it worksGet access
Device Software Anomaly Triage Agent: what goes in, what the agent does and what you get

What it does

Software anomalies on a medical device are often logged as plain bugs, with no severity and no check against the hazard analysis. This agent reads each new anomaly in the tracker, classifies severity using the team's scheme, and looks up linked hazards in the risk file. If the anomaly could touch a hazard, it flags the item for safety review and asks for reproduction steps, logs and software version. When new information arrives it reclassifies and rechecks the hazard link. It also lists anomalies that are open against a planned release. The engineer approves the severity and the release decision. Edge case: an intermittent display glitch has no reproduction yet, so the agent keeps it at the higher provisional severity until evidence says otherwise.

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 New anomaly logged in the tracker 2 USES A TOOL Read the anomaly, logs and software version 3 DOES Propose a severity using the team scheme 4 USES A TOOL Search the risk file for linked hazards 5 CHECKS THE RESULT Is a linked hazard found or ruled out with a reason? If not: widen the search to related functions andrequest input from the engineer. Back to step 4. 6 DOES Flag items needing safety review 7 USES A TOOL Request reproduction steps and logs from thereporter 8 CHECKS THE RESULT Is the anomaly reproduced or explained? If not: keep the higher provisional severity and set afollow-up date. Back to step 3. 9 DOES Build the release impact list 10 YOU APPROVE Engineer approves severity and release decision 11 RESULT Triage report and updated tracker
Read the steps as a list
  1. New anomaly logged in the tracker
  2. Read the anomaly, logs and software version
  3. Propose a severity using the team scheme
  4. Search the risk file for linked hazards
  5. Is a linked hazard found or ruled out with a reason?If not: widen the search to related functions and request input from the engineer. Back to step 4.
  6. Flag items needing safety review
  7. Request reproduction steps and logs from the reporter
  8. Is the anomaly reproduced or explained?If not: keep the higher provisional severity and set a follow-up date. Back to step 3.
  9. Build the release impact list
  10. Engineer approves severity and release decisionThe agent waits here for your OK.
  11. Triage report and updated tracker

How it decides

Severity follows the worst credible harm in the risk file. Missing evidence keeps the higher provisional severity.

  • Classify by worst credible harm, not by how often it happens
  • Keep provisional severity when not reproduced
  • Send any anomaly with a hazard link to safety review
  • Block listing as fixed until a verification record is linked

Make it yours

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

  • Severity scheme and levels
  • Hazard source file location
  • Days before a follow-up request (default 3)
  • Releases to monitor

What keeps you in control

It always asks you first

  • Severity classification
  • Release decision for open anomalies

Hard limits

  • Never downgrades severity without evidence
  • Never closes an anomaly

It stops when

  • Done: all open anomalies triaged and reviewed
  • Stop: risk file is unavailable

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 happensAnomaly 318 reports the pump screen freezing for 4 seconds in version 2.4.1. The agent proposed medium severity, then found hazard H-22 for delayed alarm display, so the hazard check raised it to high and flagged safety review. It asked for logs, got them next day and confirmed a memory spike. The engineer approved high severity and held the release.

More agents for biomedical engineers

Communication · Automation

Product Requirement Change Impact Agent

Every requirement change reaches all linked tests, docs and teams.

7 steps1 check3 jobs
Analysis · Decisions

User Test Prototype Iteration Agent

Turn test observations into ranked problems and verified changes for the next prototype.

11 steps2 checks8 jobs
Project tracking · Administration

Lab Meeting Action Follow-up Agent

Make sure every action item from a meeting has an owner, a date and a recorded outcome.

10 steps2 checks7 jobs
Risk management · Quality control

Usability Study Use Error Tracking Agent

Every observed use error, close call and difficulty is traced to a hazard and a mitigation before the report.

11 steps2 checks2 jobs