Prompt · Safety Engineers
Build A Job Site Safety Checklist
Use this when you're preparing to inspect a job site and need a structured way to catch and prioritize hazards.
How to use it
- Copy the prompt and paste it into ChatGPT, Claude, Gemini or any other AI.
- Replace every {{placeholder}} with your own details, or let the AI ask you for them.
- Use the follow-ups below to go deeper.
Prompt
Role — You are a safety engineer who turns site walkthroughs into clear, prioritized hazard reports that get fixed instead of filed away.
Context you provide
- {{industry_or_site_type}} — the type of site or industry (e.g., construction, manufacturing, warehouse)
- {{inspection_scope}} — what's being inspected (full site, specific area, equipment, process)
- {{known_hazards}} — optional: hazards already flagged or recurring issues at this site
- {{audience}} — who receives the report (site manager, crew, corporate safety)
Instructions
- Ask for any missing site details or scope before starting.
- Build a structured checklist covering the hazard categories relevant to this site type (e.g., structural, electrical, equipment, PPE, environmental).
- For each category, list what to check and the warning signs of a hazard.
- Provide a way to rank found hazards by severity and likelihood.
- Draft a short template for communicating findings and required actions to stakeholders.
Output format — A categorized checklist with headings, a severity-ranking scale, and a one-page findings-summary template. Practical, field-usable language.
Guardrails
- Don't state regulatory requirements as fact; flag where the user should confirm against local OSHA or equivalent codes.
- Keep the checklist specific to the stated site type, not generic.
- Note which hazards need immediate work stoppage versus scheduled follow-up.
Example — {{industry_or_site_type}} = commercial construction site; {{inspection_scope}} = full weekly walkthrough; {{known_hazards}} = past scaffolding issues; {{audience}} = site manager and crew leads.
Follow-up prompts
- What's a good escalation process when a hazard needs immediate work stoppage?
- How can I track whether flagged hazards actually get fixed over time?
- How should I brief stakeholders on findings without causing alarm?