{"id":55727,"date":"2026-09-04T09:07:02","date_gmt":"2026-09-04T09:07:02","guid":{"rendered":"https:\/\/sitemate.com\/?p=55727"},"modified":"2026-09-08T05:19:27","modified_gmt":"2026-09-08T05:19:27","slug":"how-to-use-claude-to-review-rams","status":"publish","type":"post","link":"https:\/\/sitemate.com\/au\/mcp\/use-cases\/how-to-use-claude-to-review-rams\/","title":{"rendered":"How to use Claude to review RAMS"},"content":{"rendered":"","protected":false},"excerpt":{"rendered":"","protected":false},"author":22,"featured_media":0,"comment_status":"open","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"footnotes":"","rank_math_title":"","rank_math_description":"Paste a RAMS and Claude reviews it against what a good one contains: hazards matched to controls, risk scored, a method that follows the work, a sign-on. You get a point by point list of gaps, then make the call and sign it off.","rank_math_canonical_url":"","rank_math_focus_keyword":""},"categories":[1],"tags":[],"industry":[],"product":[],"region":[145],"system":[87],"content_type":[170,173],"topic":[157],"class_list":["post-55727","post","type-post","status-publish","format-standard","hentry","category-uncategorized","region-global","system-safety","content_type-ai-landing-page","content_type-llm-use-case-page","topic-rams"],"acf":{"ai_lp_breadcrumb":"<a href=\"\/_hl-staging\/topics\/safety\/rams\/\">RAMS<\/a><span class=\"sep\">\/<\/span><span aria-current=\"page\">How to Use Claude to Review Rams<\/span>","ai_lp_h1":"How to use Claude to review RAMS","ai_lp_intro":"A principal contractor or safety lead can end up with a stack of subcontractor RAMS to check before work starts, each one 30 pages, and no easy way to see which are thorough and which are generic. Claude reads a RAMS you paste or upload and gives you a structured review: what's missing, where a hazard has no control, where the method is vague. You still own the sign-off, but you get through the stack faster and catch the gaps that matter.","ai_lp_hero_media":"<div class=\"vimeo-container ailp-video\" data-ailp-video-id=\"1195268867\" data-ailp-video-hash=\"1d30f26dc7\" data-ailp-video-title=\"How to use Claude to review RAMS\" style=\"padding:56.25% 0 0 0;position:relative;\"><\/div>","ai_lp_body":"<div class=\"ailp-block ailp-answer\">\n<p class=\"eyebrow-sm\">The short answer<\/p>\n<p>Paste or upload a RAMS and ask Claude to review it against what a good one contains: hazards matched to controls, a risk score on each, a method that follows the work, and a sign-on. It returns a point by point list of gaps and weak spots, not a pass or fail. You read its findings, make the call on each, and a competent person still signs the RAMS off.<\/p>\n<\/div>\n\n<div class=\"ailp-block ailp-heading-body\">\n<h2>How to get Claude reviewing your RAMS<\/h2>\n<p>A useful review comes down to what you ask for. Give Claude the RAMS and the standard to check it against, ask for specific findings rather than a verdict, then act on the list. Do it once and you have a checking prompt you reuse on every subcontractor RAMS that lands.<\/p>\n<\/div>\n\n<div class=\"ailp-block ailp-step-list\">\n<ol class=\"ailp-steps\">\n  <li>\n    <p class=\"eyebrow-sm\">Step 1<\/p>\n    <h3>Give Claude the RAMS and the standard<\/h3>\n    <p>Paste or upload the RAMS, and tell Claude what to hold it to: the hazards the task should cover, whether each has a control, whether risk is scored, and whether the method follows the actual work. The clearer the standard, the sharper the review.<\/p>\n    <ul class=\"checklist-tick\">\n      <li>The RAMS itself, as text or a file<\/li>\n      <li>The task it is meant to cover<\/li>\n      <li>The points that matter to you: coverage, controls, scoring, method<\/li>\n      <li>Your own template or client standard, if you have one<\/li>\n    <\/ul>\n  <\/li>\n  <li>\n    <p class=\"eyebrow-sm\">Step 2<\/p>\n    <h3>Ask for findings, not a verdict<\/h3>\n    <p>Left open, Claude says the RAMS looks fine. Ask it for a list instead: each hazard with no control, each control with no method step, anything generic, any gap against the task. A verdict hides the detail you need to fix; a list is something you can act on.<\/p>\n    <ul class=\"checklist-dot\">\n      <li>Hazards named but not controlled<\/li>\n      <li>Controls with no matching method step<\/li>\n      <li>Generic text that could be any job<\/li>\n      <li>A risk score missing or not justified<\/li>\n      <li>A missing sign-on or approval<\/li>\n    <\/ul>\n  <\/li>\n  <li>\n    <p class=\"eyebrow-sm\">Step 3<\/p>\n    <h3>Make the call on each finding<\/h3>\n    <p>The review is a first pass, not the sign-off. Read each finding against the site and the task, keep the ones that matter, and send the real gaps back to whoever wrote the RAMS. A competent person still signs it off, now with the weak spots caught.<\/p>\n  <\/li>\n<\/ol>\n<\/div>\n\n<div class=\"ailp-block ailp-heading-body\">\n<h2>Prompt guidance for reviewing RAMS with Claude<\/h2>\n<p>The prompt is what turns a soft read into a real review. Ask for specific findings against a clear standard, and reuse the prompt on every RAMS that lands, changing only the task it is meant to cover.<\/p>\n<ul class=\"checklist-dot\">\n  <li>&quot;Review this RAMS for [task]. List every hazard that has no control, and every control with no method step.&quot;<\/li>\n  <li>&quot;Check the risk scoring: is there a score and residual risk against each hazard, and is it consistent?&quot;<\/li>\n  <li>&quot;Flag anything generic that could apply to any job rather than this task.&quot;<\/li>\n  <li>&quot;Compare this RAMS against our template and list what is missing.&quot;<\/li>\n  <li>&quot;List the three weakest points I should send back before I sign it off.&quot;<\/li>\n<\/ul>\n<\/div>\n\n<div class=\"ailp-block ailp-heading-body\">\n<h2>Troubleshooting prompts for reviewing RAMS<\/h2>\n<p>A soft review is usually a soft prompt, not a limit of the assistant. These are the fixes worth trying, in order.<\/p>\n<ul class=\"checklist-dot\">\n  <li>The review is vague: you asked if it is fine. Ask for a list of specific gaps instead<\/li>\n  <li>It misses obvious gaps: name the task, so it knows what the RAMS should have covered<\/li>\n  <li>It rewrites instead of reviewing: tell it to report findings, not to edit the document<\/li>\n  <li>It is too soft: ask it to assume a principal contractor will reject anything generic<\/li>\n  <li>It can't read the file: paste the RAMS as text if the upload doesn't come through<\/li>\n<\/ul>\n<\/div>\n\n<div class=\"ailp-block ailp-heading-body\">\n<h2>Limitations<\/h2>\n<p>Claude reviews the RAMS. It doesn't sign it off, and it reviews the document, not the site. A review that clears a RAMS reading well but blind to the site is worse than none, so the call is still yours.<\/p>\n<\/div>\n\n<div class=\"ailp-block ailp-subheading-body\">\n<h3>What Claude does well here<\/h3>\n<p>Given the RAMS and a clear standard, this is what it does well.<\/p>\n<ul class=\"checklist-tick\">\n  <li>Read a full RAMS and check it against what a good one contains<\/li>\n  <li>Flag hazards with no control, and controls with no method step<\/li>\n  <li>Spot generic text that doesn't fit the task<\/li>\n  <li>Check risk scoring is present and consistent<\/li>\n  <li>Compare a RAMS against your own template or standard<\/li>\n<\/ul>\n<\/div>\n\n<div class=\"ailp-block ailp-subheading-body\">\n<h3>What it does not do<\/h3>\n<p>And here's what it doesn't do, so the review never stands in for the sign-off.<\/p>\n<ul class=\"checklist-cross\">\n  <li>Know the site. It reviews the document, not the work<\/li>\n  <li>Decide whether a control is adequate. A competent person owns that<\/li>\n  <li>Sign the RAMS off or make it compliant<\/li>\n  <li>Guarantee it caught everything. Your review still runs<\/li>\n<\/ul>\n<\/div>\n\n<div class=\"ailp-block ailp-heading-body\">\n<h2>What this assumes you already do<\/h2>\n<p>This page assumes you already collect and check RAMS from subcontractors. Claude speeds the checking up, but it only helps if someone owns reviewing them and signing off. None of this is a setup step, it's the practice the review sits on.<\/p>\n<ul class=\"checklist-tick\">\n  <li>You collect a RAMS from each subcontractor before work<\/li>\n  <li>Someone owns reviewing them<\/li>\n  <li>A competent person signs off the method<\/li>\n  <li>You know the task well enough to tell a thorough RAMS from a generic one<\/li>\n<\/ul>\n<\/div>\n\n<div class=\"ailp-block ailp-heading-body\">\n<h2>Why use Claude to review RAMS<\/h2>\n<p>Checking one RAMS properly takes time. Checking twenty before a Monday start, each one 30 pages, is where it falls down. So the stack gets a quick skim, the generic ones slip through, and the liability sits with whoever signed them off. A near miss later traces back to a control that was never in the method.<\/p>\n<p>A structured first pass in minutes changes that. Claude reads each RAMS against what a good one contains and hands you the gaps, so your time goes on the calls that matter rather than reading every page cold. The stack clears, and the weak ones get caught before work starts.<\/p>\n<\/div>\n\n<div class=\"ailp-block ailp-heading-body\">\n<h2>What a good RAMS review catches<\/h2>\n<p>A review worth running finds the things a quick read misses:<\/p>\n<ul class=\"checklist-tick\">\n  <li>Hazards named but never controlled<\/li>\n  <li>Controls with no matching method step<\/li>\n  <li>Generic text that fits any job, not this one<\/li>\n  <li>Risk scoring missing or inconsistent<\/li>\n  <li>A missing sign-on or approval<\/li>\n<\/ul>\n<\/div>\n\n<div class=\"ailp-block ailp-callout\">\n<p>It's Sunday night, six subcontractor RAMS are in your inbox for a Monday start, and you're the name that signs them off. Skim them and the generic ones slip through. Run each through a proper check first, and the gaps show up before the crew does.<\/p>\n<\/div>\n\n<div class=\"ailp-block ailp-chart\">\n<h2>What people use AI assistants for inside Sitemate<\/h2>\n<p>Every named job customers ran through an AI assistant against their own Sitemate data, over six months.<\/p>\n<figure class=\"ailp-chart-figure ailp-chart-bar\" data-ailp-chart data-ailp-style=\"bar\">\n<table class=\"ailp-chart-table\">\n<caption class=\"ailp-chart-caption\">What people use AI assistants for inside Sitemate<\/caption>\n<thead>\n<tr><th scope=\"col\">The job<\/th><th scope=\"col\">Share of tasks<\/th><\/tr>\n<\/thead>\n<tbody>\n<tr><th scope=\"row\">Users and access<\/th><td data-ailp-value=\"35\">35%<\/td><\/tr>\n<tr><th scope=\"row\">Site diaries<\/th><td data-ailp-value=\"14\">14%<\/td><\/tr>\n<tr><th scope=\"row\">Timesheets and attendance<\/th><td data-ailp-value=\"11\">11%<\/td><\/tr>\n<tr><th scope=\"row\">Plant, equipment and fleet<\/th><td data-ailp-value=\"9\">9%<\/td><\/tr>\n<tr><th scope=\"row\">Invoicing and commercial<\/th><td data-ailp-value=\"9\">9%<\/td><\/tr>\n<tr><th scope=\"row\">Inspections and checklists<\/th><td data-ailp-value=\"8\">8%<\/td><\/tr>\n<tr><th scope=\"row\">Compliance and audit<\/th><td data-ailp-value=\"6\">6%<\/td><\/tr>\n<tr><th scope=\"row\">Prestart and plant checks<\/th><td data-ailp-value=\"6\">6%<\/td><\/tr>\n<tr><th scope=\"row\">Incidents and hazards<\/th><td data-ailp-value=\"4\">4%<\/td><\/tr>\n<tr><th scope=\"row\">Scheduled automation<\/th><td data-ailp-value=\"4\">4%<\/td><\/tr>\n<\/tbody>\n<\/table>\n<figcaption class=\"ailp-chart-source\">Source: Sitemate MCP usage study, 3,055 unique tasks, Mar to Aug 2026<\/figcaption>\n<\/figure>\n<\/div>\n\n<div class=\"ailp-block ailp-subheading-body\">\n<h3>Teams reach for the assistant on the documents they repeat<\/h3>\n<p>The jobs people run through an assistant are the ones that recur: diaries, timesheets, inspections, the structured documents produced and checked again and again. A RAMS is that shape of task exactly, which is why reviewing one is where an assistant earns its place.<\/p>\n<\/div>\n\n<div class=\"ailp-block ailp-heading-body\">\n<h2>Connecting Sitemate MCP to Claude<\/h2>\n<p>Three steps, and the first one isn't in Claude. An Org Controller has to switch MCP access on in Dashpivot by Sitemate before the connector will authenticate for anyone.<\/p>\n<\/div>\n\n<div class=\"ailp-block ailp-step-list\">\n<ol class=\"ailp-steps\">\n  <li>\n    <p class=\"eyebrow-sm\">Step 1<\/p>\n    <h3>Turn on MCP access in Dashpivot<\/h3>\n    <p>An Org Controller opens Settings, then Security, and sets the MCP Access toggle to anything other than None. Save changes. Until this is on, the connector won't authenticate.<\/p>\n  <\/li>\n  <li>\n    <p class=\"eyebrow-sm\">Step 2<\/p>\n    <h3>Add the Sitemate connector in Claude<\/h3>\n    <p>In Claude, go to Settings, then Connectors, then Add and Browse connectors. Search for Sitemate, pick the connector tagged Community, and click Connect. Sign in with your Sitemate credentials and approve access when prompted.<\/p>\n  <\/li>\n  <li>\n    <p class=\"eyebrow-sm\">Step 3<\/p>\n    <h3>Pick it in a chat<\/h3>\n    <p>Start a new chat, click the plus button, open Connectors and select Sitemate. What you can read is whatever your Dashpivot permissions already allow, so connecting Claude doesn't widen anyone's access.<\/p>\n  <\/li>\n<\/ol>\n<\/div>\n\n<div class=\"ailp-block ailp-subheading-body\">\n<h3>If the connector will not appear<\/h3>\n<p>Three things account for almost every failed connection, and the first is the most common by a distance.<\/p>\n<ul class=\"checklist-cross\">\n  <li>Nothing connects at all: MCP Access is still off in Dashpivot, and only an Org Controller can turn it on<\/li>\n  <li>Sitemate isn't in the connector list: search &quot;Sitemate&quot; exactly, and look for the Community tag<\/li>\n  <li>The connector doesn't show in a chat: start a new chat, click the plus button, and pick it under Connectors<\/li>\n<\/ul>\n<\/div>\n\n<div class=\"ailp-block ailp-heading-body\">\n<h2>Where the RAMS and its review live<\/h2>\n<p>The RAMS you reviewed still has to be run, signed and kept, and the review is worth more when it sits with the record. Here's where both go.<\/p>\n<\/div>\n\n<div class=\"ailp-block ailp-step-list\">\n<ol class=\"ailp-steps\">\n  <li>\n    <h3>Storm captures it on site<\/h3>\n    <p>The RAMS gets run and signed on a phone or tablet while the crew is standing there, with the sign-on attached, rather than a document emailed round and chased later.<\/p>\n  <\/li>\n  <li>\n    <h3>Dashpivot holds it as the record<\/h3>\n    <p>Each RAMS and its sign-on lands against its job as a structured record, versioned and audit ready, so the review you did stays attached to the document it checked.<\/p>\n  <\/li>\n  <li>\n    <h3>Sitemate MCP reads it back<\/h3>\n    <p>With the RAMS captured, Claude can read across them: which jobs have a signed RAMS, which are still open, where the same gap keeps recurring. Your own permissions still apply.<\/p>\n  <\/li>\n<\/ol>\n<\/div>","ai_lp_body_2":"","ai_lp_comparison":"","ai_lp_band_logo":"<img class=\"product-logo\" src=\"https:\/\/sitemate-headless.pages.dev\/shared\/assets\/logos\/dashpivot-logo.svg\" alt=\"Dashpivot\">","ai_lp_band_heading":"","ai_lp_band_intro":"Claude reviews the RAMS, but it still has to be run and signed. Storm captures the RAMS and its sign-on as it happens on site, so the signed record sits alongside the review.","ai_lp_capabilities":"<div class=\"custom-col-12 bento-col\">\n  <div class=\"custom-grid\">\n    <div class=\"custom-col-6 ailp-cap ailp-cap--split\">\n      <div class=\"ailp-cap-icon\"><img src=\"\/wp-content\/uploads\/2026\/08\/ai-storm-icon.svg\" alt=\"\"><\/div>\n      <div class=\"ailp-cap-text\">\n        <h3>AI Form Fill<\/h3>\n        <p>Capture what you see and say, and Storm fills the form for you. The details land in the right fields automatically, so a complete record exists the moment the work is done.<\/p>\n      <div class=\"button-group\">\n        <a class=\"text-buttons\" href=\"\/storm\/\">Learn more about Storm<img class=\"btn-arrow\" src=\"https:\/\/sitemate-headless.pages.dev\/shared\/assets\/icons\/grey-arrow.svg\" alt=\"\" width=\"16\" height=\"16\"><\/a>\n      <\/div>\n      <\/div>\n    <\/div>\n    <div class=\"custom-col-6\">\n      <img class=\"ailp-media\" src=\"\/wp-content\/uploads\/2026\/08\/storm-form-fill.webp\" alt=\"AI Form Fill\">\n    <\/div>\n  <\/div>\n<\/div>\n\n<div class=\"custom-col-4 bento-col ailp-cap\">\n  <div class=\"ailp-cap-icon\">\n    <div class=\"sm-icon-wrap gradient-light\"><img src=\"\/wp-content\/uploads\/2026\/08\/storm-listens-to-the-work.svg\" alt=\"\"><\/div>\n  <\/div>\n  <div class=\"ailp-cap-text\">\n    <h3>Listens to the work<\/h3>\n    <p>Captures voice cleanly through machinery, wind, and movement. Accents and trade language understood. No headset needed.<\/p>\n<div class=\"button-group\">\n  <a class=\"text-buttons\" href=\"\/storm\/\">Learn more about Storm<img class=\"btn-arrow\" src=\"https:\/\/sitemate-headless.pages.dev\/shared\/assets\/icons\/grey-arrow.svg\" alt=\"\" width=\"16\" height=\"16\"><\/a>\n<\/div>\n  <\/div>\n<\/div>\n\n<div class=\"custom-col-4 bento-col ailp-cap ailp-cap--flip\">\n  <div class=\"ailp-cap-icon\">\n    <div class=\"sm-icon-wrap gradient-light\"><img src=\"\/wp-content\/uploads\/2026\/08\/storm-works-on-site.svg\" alt=\"\"><\/div>\n  <\/div>\n  <div class=\"ailp-cap-text\">\n    <h3>Works on site<\/h3>\n    <p>Built for gloves, cabs, and screen protectors. Captures inputs offline and processes in the background. No waiting on the field.<\/p>\n<div class=\"button-group\">\n  <a class=\"text-buttons\" href=\"\/storm\/\">Learn more about Storm<img class=\"btn-arrow\" src=\"https:\/\/sitemate-headless.pages.dev\/shared\/assets\/icons\/grey-arrow.svg\" alt=\"\" width=\"16\" height=\"16\"><\/a>\n<\/div>\n  <\/div>\n<\/div>\n\n<div class=\"custom-col-4 bento-col ailp-cap\">\n  <div class=\"ailp-cap-icon\">\n    <div class=\"sm-icon-wrap gradient-light\"><img src=\"\/wp-content\/uploads\/2026\/08\/storm-you-stay-in-control.svg\" alt=\"\"><\/div>\n  <\/div>\n  <div class=\"ailp-cap-text\">\n    <h3>You stay in control<\/h3>\n    <p>Storm never commits a record on its own. Every output is a draft for human review. Admins control what Storm can write to.<\/p>\n<div class=\"button-group\">\n  <a class=\"text-buttons\" href=\"\/storm\/\">Learn more about Storm<img class=\"btn-arrow\" src=\"https:\/\/sitemate-headless.pages.dev\/shared\/assets\/icons\/grey-arrow.svg\" alt=\"\" width=\"16\" height=\"16\"><\/a>\n<\/div>\n  <\/div>\n<\/div>","ai_lp_faq_heading":"Frequently Asked Questions","ai_lp_faq":[{"question":"Can Claude sign off a RAMS as compliant?","content":"<p>No. It reviews the document and flags gaps. Whether a control is adequate, and whether the RAMS is signed off, is a competent person's call. The review makes that call better informed, it doesn't replace it.<\/p>"},{"question":"Does it read the site, or just the document?","content":"<p>Just the document. It checks what the RAMS says against what a good one should contain. It doesn't know what's actually on site, so your own read of the ground still runs.<\/p>"},{"question":"Can I check a RAMS against our own standard?","content":"<p>Yes. Give Claude your template or client standard alongside the RAMS and ask it to list what's missing against it, rather than reviewing in the abstract.<\/p>"},{"question":"Will it catch everything?","content":"<p>No, and you shouldn't run it as if it will. It's a fast first pass that catches the common gaps, hazards with no control, generic text, missing scoring. Your review still runs on top of it.<\/p>"},{"question":"Does this work with the other assistants as well?","content":"<p>Yes. The same approach holds whichever assistant your team uses, and the Sitemate connection supports Claude, ChatGPT and Microsoft Copilot.<\/p>"},{"question":"Where does the reviewed RAMS end up?","content":"<p>Run and signed on site, then stored in Dashpivot as the record with Storm. The review sits alongside the RAMS it checked, so the gaps you found and the signed document stay together.<\/p>"}],"ai_lp_version_history":"","ai_lp_closing_heading":"","ai_lp_closing_body":""},"custom_permalink":"mcp\/use-cases\/how-to-use-claude-to-review-rams","_links":{"self":[{"href":"https:\/\/sitemate.com\/au\/wp-json\/wp\/v2\/posts\/55727","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/sitemate.com\/au\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/sitemate.com\/au\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/sitemate.com\/au\/wp-json\/wp\/v2\/users\/22"}],"replies":[{"embeddable":true,"href":"https:\/\/sitemate.com\/au\/wp-json\/wp\/v2\/comments?post=55727"}],"version-history":[{"count":1,"href":"https:\/\/sitemate.com\/au\/wp-json\/wp\/v2\/posts\/55727\/revisions"}],"predecessor-version":[{"id":55728,"href":"https:\/\/sitemate.com\/au\/wp-json\/wp\/v2\/posts\/55727\/revisions\/55728"}],"wp:attachment":[{"href":"https:\/\/sitemate.com\/au\/wp-json\/wp\/v2\/media?parent=55727"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/sitemate.com\/au\/wp-json\/wp\/v2\/categories?post=55727"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/sitemate.com\/au\/wp-json\/wp\/v2\/tags?post=55727"},{"taxonomy":"industry","embeddable":true,"href":"https:\/\/sitemate.com\/au\/wp-json\/wp\/v2\/industry?post=55727"},{"taxonomy":"product","embeddable":true,"href":"https:\/\/sitemate.com\/au\/wp-json\/wp\/v2\/product?post=55727"},{"taxonomy":"region","embeddable":true,"href":"https:\/\/sitemate.com\/au\/wp-json\/wp\/v2\/region?post=55727"},{"taxonomy":"system","embeddable":true,"href":"https:\/\/sitemate.com\/au\/wp-json\/wp\/v2\/system?post=55727"},{"taxonomy":"content_type","embeddable":true,"href":"https:\/\/sitemate.com\/au\/wp-json\/wp\/v2\/content_type?post=55727"},{"taxonomy":"topic","embeddable":true,"href":"https:\/\/sitemate.com\/au\/wp-json\/wp\/v2\/topic?post=55727"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}