@elitedcs/ghl-mcp 3.83.4 → 3.83.6

package/CHANGELOG.md

@@ -2,6 +2,57 @@
 
 ## Unreleased
 
+## 3.83.6 — what a design run actually costs you, measured
+
+3.83.5 gave a running job a clock. This gives it an honest number on it.
+
+How long a run takes was written by hand in three places and every one was wrong.
+The brand page said a deck takes "roughly ten minutes". The run API said "twenty
+minutes or more". The clock added in 3.83.5 said "ten to twenty". Nobody had ever
+timed one.
+
+Two runs have now been measured. A six-slide deck took 12 minutes 39 seconds and
+$1.74 for a single pass. A conversion page on a real brand took 26 minutes and
+$3.75 across three. So a pass is 9 to 13 minutes and $1.25 to $1.75, and a run may
+use three of them.
+
+That sentence now lives in one file and every screen reads it, so the three cannot
+drift apart again.
+
+**The cost is now stated before you press the button, not discovered afterwards.**
+These runs use your own Claude, never ours. On a Max plan that is usage rather than
+a bill, which is exactly why the figure belongs on the screen: on pay as you go it
+is real money, and you should know roughly what a button costs before you press it.
+
+## 3.83.5 — the brand page's buttons work, and a run tells you it is working
+
+If you use Design Studio, every button on a brand's own page has been dead since
+3.78.0. "Make it" did nothing but load a blank page. So did "Save changes", which
+means any accent, font, name or logo you edited there was thrown away without
+saying so. Nine days, five releases. One piece of broken punctuation in the page's
+script switched all of it off at once, and because a browser discards a script like
+that in silence, nothing on screen ever said so.
+
+Both are fixed, and a new check now reads every page the cockpit serves and refuses
+to ship one whose script will not run.
+
+A design run also used to tell you nothing at all while it worked. Claude writes the
+whole file in a single answer, so the log stayed empty for ten or twenty minutes,
+and there was no way to tell a working run from a dead one. The sensible thing to do
+was stop it, which was the wrong thing to do. The run now shows a clock, counts the
+passes, and says plainly that nothing will appear until a pass finishes.
+
+While reading what gets sent to Claude on those runs, we found the brief was asking
+for two tools that are deliberately switched off for them, and then telling Claude
+in the next line that it had no tools. It no longer does that.
+
+**A correction to 3.83.4.** That release said a form field written without its id,
+key and data type "is stored and shows as raw data instead of a rendered field."
+That is wrong, and we tested it after we shipped it rather than before. A field
+without those still renders normally in the form editor. What they actually do is
+tie the answer to the contact record, which is reason enough to set them, and
+`update_form` now says only that.
+
 ## 3.83.4 — one door for every write, so "protected" cannot quietly stop meaning protected
 
 3.83.3 fixed the holes. This closes the shape that kept making them.
@@ -28,6 +79,11 @@ RENDER rather than only how to send one. A custom question needs the real custom
 field's id, key and data type from your account; a field written without them is
 stored and shows as raw data instead of a rendered field.
 
+> **Corrected 2026-09-12, in 3.83.5.** The last sentence is wrong. It was tested
+> after this release shipped, not before: a field without those keys still renders
+> normally in the form editor. They tie the answer to the contact record. That is
+> what `update_form` says now.
+
 ## 3.83.3 — an account you marked protected is now actually protected
 
 **Marking a client account protected did not stop fourteen write paths from writing to it.**

package/README.md

@@ -897,7 +897,7 @@ Built by **[Elite DCs, LLC](https://elitedcs.com)** — a digital marketing and
 
 **Tech stack:** TypeScript, Node.js, esbuild, MCP SDK, Zod, GHL API v2, Firebase Auth
 
-**Version:** 3.83.4
+**Version:** 3.83.6
 
 ---
 

package/dist/index.js

@@ -17177,7 +17177,7 @@ function registerFormBuilderTools(server2, builderClient, publicClient) {
   );
   server2.tool(
     "update_form",
-    "Update a form's structure \u2014 fields, labels, conditional logic, auto-responder, email notifications, styling. Use get_form_full first to read the current form, modify the formData object, then pass it here. If name is omitted, the current name is preserved (one extra GET). IMPORTANT \u2014 a NEW field will not render in GoHighLevel's form editor unless it carries the binding GHL needs: a custom-question field must include the real custom field's `id`, `fieldKey` and `dataType` from THIS account (read them with get_custom_fields first; do not invent them), and a standard field (first_name, last_name, email, phone) needs `standard: true` with its `tag`. A field written without that binding is stored but shows as raw data instead of a rendered field. These shapes were reverse-engineered from live UI-built forms; copy an existing field from get_form_full rather than composing one from memory.",
+    "Update a form's structure \u2014 fields, labels, conditional logic, auto-responder, email notifications, styling. Use get_form_full first to read the current form, modify the formData object, then pass it here. If name is omitted, the current name is preserved (one extra GET). IMPORTANT \u2014 a custom-question field should carry the real custom field's `id`, `fieldKey` and `dataType` from THIS account (read them with get_custom_fields; do not invent them), and a standard field (first_name, last_name, email, phone) needs `standard: true` with its `tag`. That binding is what ties the answer to the contact record. A field WITHOUT it still renders in the form editor \u2014 verified live on 2026-09-12, so do not expect a missing binding to show up as a broken field. Copy an existing field from get_form_full rather than composing one from memory; these shapes were reverse-engineered from live UI-built forms.",
     {
       formId: import_zod44.z.string().describe("The form ID to update."),
       formData: import_zod44.z.record(import_zod44.z.string(), import_zod44.z.unknown()).describe("The updated formData object containing form fields, settings, autoResponder config, etc. Fields live at formData.form.fields; each one needs the binding described above or the editor cannot render it."),
@@ -21368,7 +21368,7 @@ function tokensCss(d, theme) {
   ];
   return [...light, "", ...dark].join("\n");
 }
-function composeBrief(d, brief) {
+function composeBrief(d, brief, packOpts = {}) {
   const conversionMode = brief.options.pageMode === "conversion" && brief.conversion;
   const { design, antiDefault } = rulesFor(brief.medium, "html", conversionMode ? "conversion" : "standard");
   const opts = brief.options;
@@ -21485,7 +21485,7 @@ ${cleanLogo.slice(0, 6e3)}` : d.logo?.dataUri ? `Logo markup: <img src="${d.logo
     "## Before you finish",
     ...checklist.map((c) => `- [ ] ${c}`),
     "",
-    `Write the whole file to \`${fileName}\`. Then run lint_design_contract, then verify_design; fix what they name; stop after three passes and report.`
+    ...packOpts.tools === false ? [] : [`Write the whole file to \`${fileName}\`. Then run lint_design_contract, then verify_design; fix what they name; stop after three passes and report.`]
   ].join("\n");
   return { fileName, markdown, tokensCss: css2, rules: design, antiDefault, contract: chosenContract, checklist, bannedTells };
 }
@@ -27187,6 +27187,20 @@ var init_run_jobs = __esm({
   }
 });
 
+// src/command-os/design/run-expectations.ts
+var RUN_MAX_PASSES, RUN_PASS_MINUTES, RUN_PASS_COST, RUN_TOTAL_MINUTES, RUN_TOTAL_COST, RUN_EXPECTATION;
+var init_run_expectations = __esm({
+  "src/command-os/design/run-expectations.ts"() {
+    "use strict";
+    RUN_MAX_PASSES = 3;
+    RUN_PASS_MINUTES = "9 to 13 minutes";
+    RUN_PASS_COST = "$1.25 to $1.75";
+    RUN_TOTAL_MINUTES = "25 to 40 minutes";
+    RUN_TOTAL_COST = "about $5";
+    RUN_EXPECTATION = `Each pass takes ${RUN_PASS_MINUTES} and costs ${RUN_PASS_COST} of your own Claude usage, and a run may take up to ${RUN_MAX_PASSES} passes. Allow ${RUN_TOTAL_MINUTES} and up to ${RUN_TOTAL_COST}.`;
+  }
+});
+
 // src/run-cost.ts
 function runCostPath() {
   return path18.join(appDataDir(), "command-os-run-cost.jsonl");
@@ -28591,7 +28605,7 @@ var require_package = __commonJS({
   "package.json"(exports2, module2) {
     module2.exports = {
       name: "@elitedcs/ghl-mcp",
-      version: "3.83.4",
+      version: "3.83.6",
       mcpName: "io.github.drjerryrelth/ghl-command",
       description: "GoHighLevel MCP Server for Claude. 250 tools \u2014 full CRM, automation, marketing control, account-wide workflow audit, live funnel-capture verification, and the only programmatic GHL workflow builder, now multi-tenant across client accounts.",
       main: "dist/index.js",
@@ -32369,7 +32383,7 @@ ${d.voice.samples.length ? `<h2>Voice</h2><ul>${d.voice.samples.map((s) => `<li>
 </div>
 ${opts.screenshot ? `<h2>Their site, as rendered</h2><img class="shot" src="/brand/${encodeURIComponent(d.slug)}/screenshot.png" alt="Screenshot of ${esc5(d.name)}'s website">` : ""}
 <h2>Make something in this brand</h2>
-<p class="small">Runs on <b>your own Claude</b>, on this computer \u2014 never our key. A deck takes roughly ten minutes, a page twenty or more; the run keeps going if you close this tab. Write the headline and the points you want made, not instructions.</p>
+<p class="small">Runs on <b>your own Claude</b>, on this computer, never our key. ${esc5(RUN_EXPECTATION)} The run keeps going if you close this tab. Write the headline and the points you want made, not instructions.</p>
 <form id="make"><div class="two">
 <div><label for="mk-medium">What</label><select id="mk-medium"><option value="deck">A deck</option><option value="site">A landing page</option><option value="report">A report</option></select></div>
 <div><label for="mk-slides">Slides (a deck only)</label><input id="mk-slides" type="number" min="1" max="40" value="6"></div>
@@ -32397,8 +32411,19 @@ ${d.provenance.source ? `<p class="small">Source: ${esc5(d.provenance.source.url
 ${opts.dir ? `<p class="small">Stored at ${esc5(opts.dir)}</p>` : ""}
 ${foot(SHELL_JS(opts.seatToken ?? "") + `
 (function(){
-var poll=null,runId=null;
-function stopPolling(){if(poll){clearInterval(poll);poll=null}}
+var poll=null,runId=null,started=0,ticker=null,passes=0;
+var EXPECT=${scriptJson(RUN_EXPECTATION)};
+function stopPolling(){if(poll){clearInterval(poll);poll=null}if(ticker){clearInterval(ticker);ticker=null}}
+function elapsed(ms){var s=Math.floor(ms/1000),m=Math.floor(s/60);s=s%60;return m?(m+"m "+(s<10?"0":"")+s+"s"):(s+"s")}
+/* The run prints NOTHING while Claude writes - it answers in one go, so the log
+   box sits still for ten minutes or more. Silence used to be indistinguishable
+   from a dead run, and the rational move was to press Stop on a working one
+   (BUG-2026-09-12-03). So the cockpit keeps its own clock: it needs nothing
+   from the child process, and it ticks every second, not every poll. */
+function tick(){
+  if(!started)return;
+  say("mk-status","Working on your Claude: "+elapsed(Date.now()-started)+" so far"+(passes?(", on pass "+(passes+1)):"")+". Claude writes the whole file in one answer, so nothing appears below until a pass finishes. "+EXPECT+" Closing this tab does not stop it.");
+}
 function show(t){var el=document.getElementById("mk-log");el.hidden=false;el.textContent=t}
 document.getElementById("make").addEventListener("submit",async function(e){
   e.preventDefault();
@@ -32415,24 +32440,25 @@ document.getElementById("make").addEventListener("submit",async function(e){
     var r=await post("/api/design/run",body);
     runId=r.id;
     document.getElementById("mk-cancel").hidden=false;
-    say("mk-status","Running on your Claude. This keeps going if you close the tab.");
+    started=Date.now();passes=0;tick();ticker=setInterval(tick,1000);
     poll=setInterval(async function(){
       var s=await (await fetch("/api/design/run?id="+encodeURIComponent(runId))).json();
-      show((s.tail||[]).join("
-"));
+      var lines=s.tail||[];
+      show(lines.join("\\n"));
+      passes=lines.filter(function(l){return l.trim().slice(0,5)==="pass "}).length;
       if(s.status!=="running"){
         stopPolling();go.disabled=false;document.getElementById("mk-cancel").hidden=true;
         var run=s.run||{};
         var design=run.ok?"design PASS on pass "+run.greenAt:"design NOT GREEN";
-        var copy=run.copy?(run.copy.error?" \xB7 copy not graded":" \xB7 copy "+run.copy.verdict+(run.copy.firstBreakdown?" (first breakdown: "+run.copy.firstBreakdown+")"):"")):"";
+        var copy=run.copy?(run.copy.error?" \xB7 copy not graded":" \xB7 copy "+run.copy.verdict+(run.copy.firstBreakdown?" (first breakdown: "+run.copy.firstBreakdown+")":"")):"";
         say("mk-status",s.status==="done"?design+copy+" \xB7 "+(run.file||s.outDir):"The run stopped: "+design+copy,s.status==="done"&&run.ok?"ok":"err");
       }
     },4000);
-  }catch(err){say("mk-status",err.message,"err");go.disabled=false}
+  }catch(err){stopPolling();started=0;say("mk-status",err.message,"err");go.disabled=false}
 });
 document.getElementById("mk-cancel").addEventListener("click",async function(){
   if(!runId)return;
-  try{await post("/api/design/cancel",{id:runId});stopPolling();say("mk-status","Stopped.");document.getElementById("mk-go").disabled=false;this.hidden=true}catch(e){say("mk-status",e.message,"err")}
+  try{await post("/api/design/cancel",{id:runId});stopPolling();started=0;say("mk-status","Stopped.");document.getElementById("mk-go").disabled=false;this.hidden=true}catch(e){say("mk-status",e.message,"err")}
 });
 })();
 document.getElementById("override").addEventListener("submit",async(e)=>{e.preventDefault();const g=(id)=>document.getElementById(id).value.trim();const body={slug:${scriptJson(d.slug)}};if(g("accent"))body.accent=g("accent");if(g("displayFont"))body.displayFont=g("displayFont");if(g("bodyFont"))body.bodyFont=g("bodyFont");if(g("brandName"))body.name=g("brandName");if(g("logoSvg"))body.logoSvg=g("logoSvg");if(Object.keys(body).length===1){say("save-status","Nothing to change.");return}const b=document.getElementById("save");b.disabled=true;try{await post("/api/brand/set",body);say("save-status","Saved.","ok");location.reload()}catch(err){say("save-status",err.message,"err");b.disabled=false}});
@@ -32442,6 +32468,7 @@ var esc5, STYLE, SHELL_JS, foot;
 var init_brand_page = __esm({
   "src/command-os/design/brand-page.ts"() {
     "use strict";
+    init_run_expectations();
     init_fonts();
     init_cockpit_shell();
     init_script_json();
@@ -36525,7 +36552,7 @@ async function runDashboard(argv) {
           }
           const job = startDesignRun(checked.value);
           res.writeHead(200, { "Content-Type": "application/json" });
-          res.end(JSON.stringify({ ok: true, id: job.id, outDir: job.outDir, note: "Started on your own Claude. This takes twenty minutes or more; poll /api/design/run?id= for progress." }));
+          res.end(JSON.stringify({ ok: true, id: job.id, outDir: job.outDir, note: `Started on your own Claude. ${RUN_EXPECTATION} Poll /api/design/run?id= for progress.` }));
         } catch (e) {
           res.writeHead(400, { "Content-Type": "application/json" });
           res.end(JSON.stringify({ ok: false, error: e.message }));
@@ -37283,6 +37310,7 @@ var init_dashboard = __esm({
     init_claude_cli();
     init_token_registry();
     init_run_jobs();
+    init_run_expectations();
     init_stage_runner();
     init_account_rules();
     init_claude_call();
@@ -43689,7 +43717,7 @@ async function runDesign(opts) {
   const medium = brief.medium;
   const outDir = (0, import_node_path7.resolve)(opts.outDir);
   (0, import_node_fs7.mkdirSync)(outDir, { recursive: true });
-  const pack = composeBrief(dna, brief);
+  const pack = composeBrief(dna, brief, { tools: false });
   const maxPasses = Math.max(1, Math.min(3, opts.maxPasses ?? 3));
   const generate = opts.generate ?? claudeGenerate(opts.model ?? DEFAULT_MODEL);
   const conversionRun = brief.options.pageMode === "conversion" && brief.conversion ? { brief: brief.conversion, gate: opts.gate ?? "ship" } : void 0;

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@elitedcs/ghl-mcp",
-  "version": "3.83.4",
+  "version": "3.83.6",
   "mcpName": "io.github.drjerryrelth/ghl-command",
   "description": "GoHighLevel MCP Server for Claude. 250 tools — full CRM, automation, marketing control, account-wide workflow audit, live funnel-capture verification, and the only programmatic GHL workflow builder, now multi-tenant across client accounts.",
   "main": "dist/index.js",