argusqa-os 9.5.0 → 9.5.1

package/README.md

@@ -79,7 +79,7 @@ The `landing/` directory contains the product landing page (React + Vite + Tailw
 
 | 🔴 Critical / 🟡 Warning / 🔵 Info | ⚙️ | 🧪 | 📋 |
 | :---: | :---: | :---: | :---: |
-| **114 distinct issue types detected** | **24 analysis engines** | **394 test assertions** | **93 test blocks** |
+| **114 distinct issue types detected** | **24 analysis engines** | **528 test assertions** | **126 test blocks** |
 
 </div>
 
@@ -632,7 +632,7 @@ Then follow with: *"Here's the context — what's causing these errors and how d
 | `npm run server` | Start the Slack slash command + interaction server (port 3001) |
 | `npm run init` | Interactive setup wizard — generates `.env` + `targets.js` |
 | `npm run test:unit` | Run 61 unit tests (no Chrome required) |
-| `npm run test:harness` | Run 93-block correctness harness (requires Chrome) |
+| `npm run test:harness` | Run 126-block correctness harness (requires Chrome) |
 
 **`npm run crawl`** — full audit of all configured routes:
 
@@ -876,7 +876,7 @@ argus/
 │   │   └── browser.js                # CdpBrowserAdapter — facade over all chrome-devtools-mcp calls
 │   ├── domain/
 │   │   └── finding.js                # createFinding() factory — canonical finding shape
-│   ├── registry.js                   # Analyzer plugin registry — registerExpensive/getCheap/getExpensive
+│   ├── registry.js                   # Analyzer plugin registry — registerCheap/registerExpensive/getCheap/getExpensive/clearAll
 │   ├── config/
 │   │   ├── targets.js                # Routes to test, thresholds, config
 │   │   └── schema.js                 # Zod validation schema; validateConfig() called inside runCrawl()
@@ -944,12 +944,12 @@ argus/
 │   └── README.md                     # Setup guide, Supabase SQL schema, env vars, deployment
 ├── scripts/
 │   └── dispatch-report.js            # Standalone Slack re-dispatch script (re-posts last report.json to Slack)
-├── test-harness/                     # Fixture server + test runner (93 blocks, 394 hard assertions, 54 fixture pages)
+├── test-harness/                     # Fixture server + test runner (126 blocks, 528 hard assertions, 53 fixture pages)
 │   ├── README.md
 │   ├── server.js                     # Express fixture server (ports 3100 dev / 3101 staging)
 │   ├── harness-config.js             # Route definitions + expected findings
-│   ├── validate.js                   # Test runner — 84 numbered blocks ([80] MCP server, [81] createFinding, [82] withRetry, [83] watch dashboard, [84] cli/init.js)
-│   ├── pages/                        # 54 fixture pages (one per detection category)
+│   ├── validate.js                   # Test runner — 126 numbered blocks ([80]–[84] MCP/createFinding/withRetry/watch/init, [85]–[93] Sprint 0.5 Tier 3, [94]–[126] gap-close Sections 1–6)
+│   ├── pages/                        # 53 fixture HTML pages (one per detection category)
 │   ├── nextjs-fixture/               # Next.js app structure for C3 discovery tests (10 files)
 │   ├── source-fixture/               # Minimal app.js for C1 codebase-analyzer tests (env var audit)
 │   └── static/
@@ -989,7 +989,7 @@ argus/
 
 ## Known MCP Tool Limitations
 
-The Chrome DevTools MCP behavioral constraints below cause **3 permanent test failures** in the harness (`391/394` pass). These are MCP-layer restrictions — they cannot be fixed in Argus code. `validate.js` now exits with code 0 when only these 3 failures remain, making the CI harness gate reliable.
+The Chrome DevTools MCP behavioral constraints below cause **3 permanent test failures** in the harness (`525/528` pass). These are MCP-layer restrictions — they cannot be fixed in Argus code. `validate.js` now exits with code 0 when only these 3 failures remain, making the CI harness gate reliable.
 
 > **`type_text` clarification**: `type_text` does fire DOM `input` events when the element is properly focused first with `mcp.click({ uid })`. Always use uid-based focus — passing `{ selector }` to `mcp.click` silently does nothing.
 

package/glama.json

@@ -1,7 +1,7 @@
 {
   "$schema": "https://glama.ai/mcp/schemas/server.json",
   "name": "argus",
-  "description": "AI-powered QA harness that audits web apps via Chrome DevTools Protocol. Catches JS errors, network failures, a11y violations, SEO issues, security headers, CSS regressions, and more — directly from Claude conversations. 6 MCP tools: argus_audit (fast 8-analyzer pass), argus_audit_full (Lighthouse + memory + responsive), argus_compare (dev vs staging diff), argus_last_report (retrieve last JSON report), argus_watch_snapshot (live tab snapshot without navigating), argus_get_context (LLM-optimized context + fix loop with snapshot_id diff). 84 test blocks, 367 hard assertions, 54 detection categories.",
+  "description": "AI-powered QA harness that audits web apps via Chrome DevTools Protocol. Catches JS errors, network failures, a11y violations, SEO issues, security headers, CSS regressions, and more — directly from Claude conversations. 6 MCP tools: argus_audit (fast 8-analyzer pass), argus_audit_full (Lighthouse + memory + responsive), argus_compare (dev vs staging diff), argus_last_report (retrieve last JSON report), argus_watch_snapshot (live tab snapshot without navigating), argus_get_context (LLM-optimized context + fix loop with snapshot_id diff). 126 test blocks, 528 hard assertions, 54 detection categories.",
   "maintainers": ["ironclawdevs27"],
   "tools": [
     {

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "argusqa-os",
-  "version": "9.5.0",
+  "version": "9.5.1",
   "mcpName": "io.github.ironclawdevs27/argus",
   "description": "Argus — AI-powered automated dev-testing platform using Chrome DevTools MCP and Claude Code",
   "keywords": [

package/src/mcp-server.js

@@ -1,6 +1,6 @@
 #!/usr/bin/env node
 /**
- * Argus MCP Server (v9.4.6)
+ * Argus MCP Server (v9.5.1)
  *
  * Exposes Argus as an MCP server so Claude (or any MCP client) can call
  * argus_audit, argus_audit_full, argus_compare, argus_last_report, and
@@ -286,7 +286,7 @@ async function handleLastReport() {
 // ── Server bootstrap ──────────────────────────────────────────────────────────
 
 const server = new Server(
-  { name: 'argus', version: '9.4.6' },
+  { name: 'argus', version: '9.5.1' },
   { capabilities: { tools: {} } },
 );
 

package/src/orchestration/orchestrator.js

@@ -424,7 +424,8 @@ export async function crawlRouteCheap(route, baseUrl, mcp) {
 
   // 0. Snapshot session-wide baselines BEFORE this route starts (D5).
   const consoleBaseline = (await browser.listConsole().catch(() => [])).length;
-  const networkBaseline = (await browser.listNetwork().catch(() => [])).length;
+  const baselineNetList = await browser.listNetwork().catch(() => []);
+  const networkMaxReqId = baselineNetList.reduce((max, r) => Math.max(max, r._reqid ?? 0), 0);
   // listConsoleRaw returns raw MCP response — normalizeArray required before .length
   const issuesBaselineRaw = await browser.listConsoleRaw({ types: ['issue'] }).catch(() => null);
   const issuesBaseline    = normalizeArray(issuesBaselineRaw).length;
@@ -468,8 +469,14 @@ export async function crawlRouteCheap(route, baseUrl, mcp) {
     });
   }
 
-  // 5. Console messages — sliced from per-route baseline
-  const consoleMsgs = (await browser.listConsole().catch(() => [])).slice(consoleBaseline);
+  // 5. Console messages — sliced from per-route baseline.
+  // Guard: chrome-devtools-mcp list_console_messages resets per navigation. If the
+  // new page has fewer total messages than the pre-navigation baseline, the baseline
+  // refers to the previous page's context and slicing by it would give an empty array.
+  // Fall back to 0 (take all messages from the new page) in that case.
+  const allConsoleMsgs = await browser.listConsole().catch(() => []);
+  const consoleSliceAt = allConsoleMsgs.length > consoleBaseline ? consoleBaseline : 0;
+  const consoleMsgs = allConsoleMsgs.slice(consoleSliceAt);
   for (const msg of consoleMsgs) {
     const text = (msg.text ?? msg.message ?? '');
     if (text.toLowerCase().includes('has been blocked by cors policy')) continue;
@@ -500,9 +507,9 @@ export async function crawlRouteCheap(route, baseUrl, mcp) {
     }
   }
 
-  // 6. Network requests — sliced from per-route baseline (cap AFTER slice, not before)
-  const networkReqs = (await browser.listNetwork())
-    .slice(networkBaseline).slice(0, 500);
+  // 6. Network requests — filtered from per-route baseline by _reqid (cap AFTER filter, not before)
+  const networkReqs = (await browser.listNetwork().catch(() => []))
+    .filter(r => (r._reqid ?? 0) > networkMaxReqId).slice(0, 500);
   for (const req of networkReqs) {
     const severity = classifyNetworkRequest(req, route.critical);
     if (severity !== null) {

package/src/orchestration/watch-mode.js

@@ -172,7 +172,7 @@ const DASHBOARD_HTML = `<!DOCTYPE html>
  * @param {number}         port         — TCP port to listen on (default 3002)
  * @returns {http.Server}
  */
-function startDashboard(getFindings, target, port) {
+export function startDashboard(getFindings, target, port) {
   const server = http.createServer((req, res) => {
     if (req.url === '/data' || req.url?.startsWith('/data?')) {
       const payload = JSON.stringify({

package/src/utils/html-reporter.js

@@ -234,7 +234,7 @@ function buildHtml(report) {
 
     ${flowSection}
 
-    <p style="text-align:center;font-size:12px;color:#9ca3af;margin-top:32px">Generated by <strong>Argus</strong> · ${esc(generatedAt)}</p>
+    <p style="text-align:center;font-size:12px;color:#9ca3af;margin-top:32px">Generated by <strong>Argus</strong> · ${esc(runDate)}</p>
   </div>
 </body>
 </html>`;

package/src/utils/mcp-parsers.js

@@ -46,7 +46,7 @@ export function parseNetworkReqResponse(raw) {
   if (typeof raw === 'object') return normalizeArray(raw);
   if (typeof raw !== 'string') return [];
   const reqs = [];
-  const re = /reqid=(\d+)\s+(\w+)\s+(\S+)\s+\[(\d+)\]/gm;
+  const re = /reqid=(\d+)\s+(\w+)\s+(\S+)\s+\[(\d+)[^\]]*\]/gm;
   let m;
   while ((m = re.exec(raw)) !== null) {
     const [, reqid, method, url, statusStr] = m;