tests: pin ink engine in _make_tui_argv npm-bootstrap tests (post-merge semantic fix)

Main's rewritten test_tui_npm_install.py tests call _make_tui_argv expecting
the Ink/npm flow unconditionally; with the dual-engine dispatch merged in,
_resolve_tui_engine() auto-selects opentui whenever ui-opentui/dist is built
in the repo, routing the call away from the path under test (first subprocess
became 'node --version' instead of 'npm run build'). Pin the engine to ink
via an autouse fixture, mirroring the existing pinning precedent in
test_tui_resume_flow.py.

scripts/build_skills_index.py

@@ -272,8 +272,9 @@ def main():
     # (well above current catalog size) lets the full catalog land in the
     # index instead of being truncated at an arbitrary build-time limit.
     SOURCE_LIMITS = {
-        # ClawHub had 49,698+ skills as of May 2026; 200k leaves headroom.
-        "clawhub": 200_000,
+        # 0 = unbounded catalog walk (max_items=0 in ClawHubSource). A positive
+        # limit bounds the walk and also enables the interactive 12s budget.
+        "clawhub": 0,
         "lobehub": 100_000,
         "browse-sh": 5_000,
         "claude-marketplace": 5_000,
@@ -297,6 +298,21 @@ def main():
     # Batch resolve GitHub paths for skills.sh entries
     all_skills = batch_resolve_paths(all_skills, auth)
 
+    # Collect which sources hit a GitHub API rate limit during the crawl.
+    # github / claude-marketplace / well-known all read api.github.com, so a
+    # rate-limited token zeroes all three at once — surfaced below so the
+    # failure message names the real cause instead of "source returned 0".
+    rate_limited_sources = {
+        name for name, source in sources.items()
+        if getattr(source, "is_rate_limited", False)
+    }
+    if rate_limited_sources:
+        print(
+            "  WARNING: GitHub API rate limit hit for: "
+            + ", ".join(sorted(rate_limited_sources)),
+            file=sys.stderr,
+        )
+
     # Deduplicate by identifier
     seen: dict[str, dict] = {}
     for skill in all_skills:
@@ -311,25 +327,9 @@ def main():
                     "browse-sh": 5, "claude-marketplace": 6, "lobehub": 7}
     deduped.sort(key=lambda s: (source_order.get(s["source"], 99), s["name"]))
 
-    # Build index
-    index = {
-        "version": INDEX_VERSION,
-        "generated_at": datetime.now(timezone.utc).isoformat(),
-        "skill_count": len(deduped),
-        "skills": deduped,
-    }
-
-    os.makedirs(os.path.dirname(OUTPUT_PATH), exist_ok=True)
-    with open(OUTPUT_PATH, "w", encoding="utf-8") as f:
-        json.dump(index, f, separators=(",", ":"), ensure_ascii=False)
-
-    elapsed = time.time() - overall_start
-    file_size = os.path.getsize(OUTPUT_PATH)
-    print(f"\nDone! {len(deduped)} skills indexed in {elapsed:.0f}s")
-    print(f"Output: {OUTPUT_PATH} ({file_size / 1024:.0f} KB)")
-
     from collections import Counter
     by_source = Counter(s["source"] for s in deduped)
+    print(f"\nCrawled {len(deduped)} skills in {time.time() - overall_start:.0f}s")
     for src, count in sorted(by_source.items(), key=lambda x: -x[1]):
         resolved = sum(1 for s in deduped
                        if s["source"] == src and s.get("resolved_github_id"))
@@ -380,14 +380,46 @@ def main():
         )
         for line in health_errors:
             print(line, file=sys.stderr)
+        if rate_limited_sources:
+            print(
+                "\nGitHub API rate limit was hit during this crawl for: "
+                + ", ".join(sorted(rate_limited_sources))
+                + ". This is the usual cause of an all-GitHub-tap collapse "
+                "(github / claude-marketplace / well-known dropping to zero "
+                "together). Re-run with a higher-quota GITHUB_TOKEN.",
+                file=sys.stderr,
+            )
         print(
             "\nIf the drop is expected (e.g. a hub is genuinely shutting "
             "down), lower the floor in scripts/build_skills_index.py "
             "EXPECTED_FLOORS in the same PR.",
             file=sys.stderr,
         )
+        # IMPORTANT: do NOT write OUTPUT_PATH on failure. The index file is
+        # gitignored, so a fresh deploy checkout has no copy on disk — leaving
+        # it absent lets website/scripts/extract-skills.py fall back to the
+        # legacy snapshot cache (or skip the unified index) instead of reading
+        # a degenerate file. Writing-then-exiting-2 was the bug that shipped an
+        # index with every GitHub-API source dropped to zero: deploy-site.yml
+        # swallows the exit code with `|| echo non-fatal`, and the partial file
+        # was already on disk for extract-skills to pick up.
         sys.exit(2)
 
+    # Healthy — only now write the index out for the docs build to consume.
+    index = {
+        "version": INDEX_VERSION,
+        "generated_at": datetime.now(timezone.utc).isoformat(),
+        "skill_count": len(deduped),
+        "skills": deduped,
+    }
+    os.makedirs(os.path.dirname(OUTPUT_PATH), exist_ok=True)
+    with open(OUTPUT_PATH, "w", encoding="utf-8") as f:
+        json.dump(index, f, separators=(",", ":"), ensure_ascii=False)
+    file_size = os.path.getsize(OUTPUT_PATH)
+    print(f"\nDone! {len(deduped)} skills indexed in "
+          f"{time.time() - overall_start:.0f}s")
+    print(f"Output: {OUTPUT_PATH} ({file_size / 1024:.0f} KB)")
+
 
 if __name__ == "__main__":
     main()

scripts/check_subprocess_stdin.py

@@ -0,0 +1,177 @@
+#!/usr/bin/env python3
+"""Check that subprocess calls in TUI-context code specify stdin=.
+
+When Hermes runs in TUI mode, the gateway child process communicates with
+the Node.js parent over a JSON-RPC protocol on stdin. Subprocess calls that
+inherit this fd can cause the gateway to exit with stdin EOF during tool
+execution (issue #14036, PR #39257).
+
+This script checks that all subprocess.run() and subprocess.Popen() calls
+in TUI-context files (agent/, tools/, plugins/, tui_gateway/) explicitly
+set stdin= to prevent fd inheritance.
+
+Exit codes:
+  0 — all calls are safe
+  1 — violations found
+  2 — script error
+
+Usage:
+  python scripts/check_subprocess_stdin.py [--fix]
+
+With --fix, prints the commands to add stdin=subprocess.DEVNULL to each
+violation (does not modify files).
+"""
+
+from __future__ import annotations
+
+import os
+import re
+import sys
+from pathlib import Path
+
+# Directories that run inside the TUI gateway child process.
+TUI_CONTEXT_DIRS = [
+    "agent/",
+    "tools/",
+    "plugins/",
+    "tui_gateway/",
+]
+
+# Files with intentional stdin= override (e.g. input= creates a pipe).
+# Format: "filepath:line" or just "filepath" to skip the whole file.
+KNOWN_SAFE = {
+    "agent/shell_hooks.py",  # uses input=stdin_json, creates a pipe
+    "plugins/security-guidance/patterns.py",  # subprocess mentions are in reminder strings, not calls
+}
+
+# Inline marker that exempts a single subprocess call from this check.
+# Put it in a comment on (or within) the call when the process MUST inherit
+# stdin — e.g. an interactive login the user explicitly invokes. Travels with
+# the line, so it survives edits that shift line numbers (unlike a pinned
+# file:line entry).
+EXEMPT_MARKER = "noqa: subprocess-stdin"
+
+# Directories to skip entirely.
+SKIP_DIRS = {
+    "tests/",
+    "scripts/",
+    "skills/",
+    "optional-skills/",
+    "hermes_cli/",
+    "gateway/",
+    "cron/",
+}
+
+
+def find_subprocess_calls(content: str, filepath: str) -> list[dict]:
+    """Find all subprocess.run/Popen calls missing stdin= in content."""
+    violations = []
+    lines = content.split("\n")
+
+    # Match only actual function calls — not comments, docstrings, or prose.
+    # The pattern requires an opening paren followed by an arg character
+    # (quote, bracket, letter, or closing paren for empty calls).
+    # This excludes ``subprocess.Popen(...)`` in docstrings and
+    # subprocess.run(...) in comments.
+    pattern = re.compile(r'subprocess\.(run|Popen)\s*\(["\'a-zA-Z_\[\(]')
+
+    for i, line in enumerate(lines):
+        # Skip comments.
+        stripped = line.lstrip()
+        if stripped.startswith("#"):
+            continue
+
+        # Skip lines where the match is inside backticks (docstring references).
+        if "``subprocess" in line:
+            continue
+
+        if not pattern.search(line):
+            continue
+
+        # Collect the full call (may span multiple lines).
+        call_start = i
+        paren_depth = 0
+        found_open = False
+        call_lines = []
+        for j in range(i, min(i + 30, len(lines))):
+            call_lines.append(lines[j])
+            for ch in lines[j]:
+                if ch == "(":
+                    paren_depth += 1
+                    found_open = True
+                elif ch == ")":
+                    paren_depth -= 1
+                    if found_open and paren_depth == 0:
+                        call_text = "\n".join(call_lines)
+
+                        # Already has stdin= → safe.
+                        if "stdin=" in call_text:
+                            break
+
+                        # Has input= → creates a pipe, safe.
+                        if "input=" in call_text:
+                            break
+
+                        # Inline exemption marker on the call itself or within
+                        # the few comment lines immediately above it → the call
+                        # intentionally inherits stdin.
+                        window_start = max(0, i - 4)
+                        preceding = "\n".join(lines[window_start:i])
+                        if EXEMPT_MARKER in call_text or EXEMPT_MARKER in preceding:
+                            break
+
+                        violations.append({
+                            "file": filepath,
+                            "line": i + 1,
+                            "snippet": line.strip()[:120],
+                        })
+                        break
+            else:
+                continue
+            break
+
+    return violations
+
+
+def main() -> int:
+    fix_mode = "--fix" in sys.argv
+    repo_root = Path(__file__).resolve().parent.parent
+    os.chdir(repo_root)
+
+    all_violations = []
+
+    for tui_dir in TUI_CONTEXT_DIRS:
+        dirpath = repo_root / tui_dir
+        if not dirpath.exists():
+            continue
+
+        for py_file in dirpath.rglob("*.py"):
+            rel = str(py_file.relative_to(repo_root))
+
+            # Skip known-safe files.
+            if rel in KNOWN_SAFE:
+                continue
+
+            # Skip test files inside tools/ etc.
+            parts = py_file.parts
+            if any(skip.rstrip("/") in parts for skip in SKIP_DIRS):
+                continue
+
+            content = py_file.read_text()
+            violations = find_subprocess_calls(content, rel)
+            all_violations.extend(violations)
+
+    if all_violations:
+        print(f"❌ {len(all_violations)} subprocess calls missing stdin=:")
+        for v in all_violations:
+            print(f"  {v['file']}:{v['line']}: {v['snippet']}")
+        if fix_mode:
+            print("\nAdd stdin=subprocess.DEVNULL to each call above.")
+        return 1
+    else:
+        print("✅ All TUI-context subprocess calls have explicit stdin=")
+        return 0
+
+
+if __name__ == "__main__":
+    sys.exit(main())

scripts/install.ps1

@@ -892,6 +892,42 @@ function Test-Node {
     return $true
 }
 
+function Update-ProcessPathForPackages {
+    # Make freshly-installed shims (rg.exe, ffmpeg.exe) visible to Get-Command in
+    # THIS process without spawning a new shell, by folding the persisted
+    # User+Machine hives plus winget's alias-shim directory into $env:Path.
+    # Called after every package-manager attempt (winget/choco/scoop): previously
+    # PATH was only refreshed inside the winget branch, so a successful
+    # choco/scoop fallback -- or any install on a box without winget -- could be
+    # misreported as "not installed".
+    #
+    # MERGE rather than overwrite: start from the existing process PATH so any
+    # process-only entries added earlier in this installer run survive, then
+    # APPEND hive/winget-Links entries not already present (case-insensitive,
+    # order-preserving dedupe). A wholesale replace would silently drop those
+    # process-only entries.
+    $candidates = @()
+    $candidates += $env:Path
+    $candidates += [Environment]::GetEnvironmentVariable("Path", "User")
+    $candidates += [Environment]::GetEnvironmentVariable("Path", "Machine")
+    $wingetLinks = Join-Path $env:LOCALAPPDATA "Microsoft\WinGet\Links"
+    if (Test-Path $wingetLinks) {
+        $candidates += $wingetLinks
+    }
+    $seen = New-Object System.Collections.Generic.HashSet[string] ([StringComparer]::OrdinalIgnoreCase)
+    $ordered = New-Object System.Collections.Generic.List[string]
+    foreach ($chunk in $candidates) {
+        if ([string]::IsNullOrEmpty($chunk)) { continue }
+        foreach ($entry in $chunk.Split(';')) {
+            $trimmed = $entry.Trim()
+            if ($trimmed -and $seen.Add($trimmed)) {
+                $ordered.Add($trimmed)
+            }
+        }
+    }
+    $env:Path = [string]::Join(';', $ordered)
+}
+
 function Install-SystemPackages {
     $script:HasRipgrep = $false
     $script:HasFfmpeg = $false
@@ -961,25 +997,33 @@ function Install-SystemPackages {
             try {
                 $output = winget install --exact --id $pkg --source winget --silent `
                     --accept-package-agreements --accept-source-agreements 2>&1
+                $code = $LASTEXITCODE
                 $output | Out-File -FilePath $log -Encoding utf8
-                "winget exit: $LASTEXITCODE" | Out-File -FilePath $log -Encoding utf8 -Append
+                "winget exit: $code" | Out-File -FilePath $log -Encoding utf8 -Append
+                # 0x8A15002B (-1978335189) = APPINSTALLER_CLI_ERROR_UPDATE_NOT_APPLICABLE.
+                # winget treats `install` on a package it already has registered as
+                # an *upgrade*, finds no newer version, and bails with this code --
+                # even when the binary is gone from disk/PATH (stale registration,
+                # files removed outside winget, or a missing alias shim). We KNOW the
+                # command was missing (that's why we're here), so a plain install
+                # dead-ends forever. Force a reinstall to repair the registration so
+                # the shim reappears.
+                if ($code -eq -1978335189) {
+                    "-> already-installed/no-upgrade; retrying with --force" | Out-File -FilePath $log -Encoding utf8 -Append
+                    $output = winget install --exact --id $pkg --source winget --silent --force `
+                        --accept-package-agreements --accept-source-agreements 2>&1
+                    $output | Out-File -FilePath $log -Encoding utf8 -Append
+                    "winget exit (force): $LASTEXITCODE" | Out-File -FilePath $log -Encoding utf8 -Append
+                }
             } catch {
                 $_ | Out-File -FilePath $log -Encoding utf8 -Append
                 "winget exit: <exception>" | Out-File -FilePath $log -Encoding utf8 -Append
             }
         }
-        # Refresh PATH from both env-var hives AND winget's alias shim directory.
-        # winget exposes packages via "command line aliases" in %LOCALAPPDATA%\
-        # Microsoft\WinGet\Links, which is added to PATH by the AppExecutionAlias
-        # machinery only in *newly-spawned* shells -- not the current process.
-        # Without this addition, Get-Command rg below would falsely return null
-        # immediately after a successful install.
-        $wingetLinks = Join-Path $env:LOCALAPPDATA "Microsoft\WinGet\Links"
-        $envPath = [Environment]::GetEnvironmentVariable("Path", "User") + ";" + [Environment]::GetEnvironmentVariable("Path", "Machine")
-        if (Test-Path $wingetLinks) {
-            $envPath = "$envPath;$wingetLinks"
-        }
-        $env:Path = $envPath
+        # Refresh PATH so packages winget exposed via "command line aliases" in
+        # %LOCALAPPDATA%\Microsoft\WinGet\Links (added to PATH only in
+        # newly-spawned shells, not this process) are visible to Get-Command below.
+        Update-ProcessPathForPackages
         if ($needRipgrep -and (Get-Command rg -ErrorAction SilentlyContinue)) {
             Write-Success "ripgrep installed"
             $script:HasRipgrep = $true
@@ -1005,6 +1049,7 @@ function Install-SystemPackages {
         foreach ($pkg in $chocoPkgs) {
             try { choco install $pkg -y 2>&1 | Out-Null } catch { }
         }
+        Update-ProcessPathForPackages
         if ($needRipgrep -and (Get-Command rg -ErrorAction SilentlyContinue)) {
             Write-Success "ripgrep installed via chocolatey"
             $script:HasRipgrep = $true
@@ -1023,6 +1068,7 @@ function Install-SystemPackages {
         foreach ($pkg in $scoopPkgs) {
             try { scoop install $pkg 2>&1 | Out-Null } catch { }
         }
+        Update-ProcessPathForPackages
         if ($needRipgrep -and (Get-Command rg -ErrorAction SilentlyContinue)) {
             Write-Success "ripgrep installed via scoop"
             $script:HasRipgrep = $true
@@ -1060,9 +1106,10 @@ function Install-Repository {
         # directory OR a symlink OR a submodule-style gitfile -- and also when
         # it's a broken stub left over from a failed previous install (e.g.
         # a partial Remove-Item that couldn't delete a locked index.lock).
-        # Validate the repo properly by asking git itself.  Two checks
-        # belt-and-braces: rev-parse AND git status.  If either fails the
-        # repo is broken and we fall through to a fresh clone.
+        # Validate the repo properly by asking git itself.  Three checks
+        # belt-and-braces: rev-parse (work tree), git status, and a resolvable
+        # HEAD (an initial commit).  If any fails the repo is broken and we
+        # fall through to a fresh clone.
         $repoValid = $false
         if (Test-Path "$InstallDir\.git") {
             Push-Location $InstallDir
@@ -1077,7 +1124,17 @@ function Install-Repository {
                 $null = & git -c windows.appendAtomically=false status --short 2>&1
                 $statusOk = ($LASTEXITCODE -eq 0)
 
-                if ($revParseOk -and $statusOk) {
+                # An interrupted previous clone leaves a repo with NO initial
+                # commit. rev-parse/status still succeed there, but the update
+                # path's `git stash` (and later `git checkout`) abort with
+                # "You do not have the initial commit yet" and fail the install
+                # (#40998). Require a resolvable HEAD so such partial checkouts
+                # are treated as broken and re-cloned fresh below.
+                $global:LASTEXITCODE = 0
+                $null = & git -c windows.appendAtomically=false rev-parse --verify HEAD 2>&1
+                $hasCommit = ($LASTEXITCODE -eq 0)
+
+                if ($revParseOk -and $statusOk -and $hasCommit) {
                     $repoValid = $true
                 }
             } catch {}
@@ -1119,7 +1176,7 @@ function Install-Repository {
                     git -c windows.appendAtomically=false stash push --include-untracked -m "$stashName"
                     if ($LASTEXITCODE -eq 0) { $autostashRef = "stash@{0}" }
                 }
-                git -c windows.appendAtomically=false fetch origin
+                git -c windows.appendAtomically=false fetch origin $Branch
                 if ($LASTEXITCODE -ne 0) { throw "git fetch failed (exit $LASTEXITCODE)" }
                 # Precedence: Commit > Tag > Branch.  Commit and Tag check
                 # out as detached HEAD intentionally -- they're meant to be
@@ -1198,16 +1255,19 @@ function Install-Repository {
             }
             $didUpdate = $true
         } else {
-            # Directory exists but isn't a usable git repo.  Wipe it and
-            # fall through to a fresh clone.  A leftover ``.git`` stub from
-            # a partial uninstall used to lock the installer into the
-            # "update" branch forever, emitting three ``fatal: not a git
-            # repository`` errors and failing with "not in a git directory".
-            Write-Warn "Existing directory at $InstallDir is not a valid git repo -- replacing it."
+            # Directory exists but isn't a usable git repo -- e.g. an
+            # interrupted clone with no initial commit (#40998), or a leftover
+            # ``.git`` stub from a partial uninstall that used to lock the
+            # installer into the "update" branch forever. Move it aside rather
+            # than deleting it -- never destroy a directory the user might still
+            # want -- and fall through to a fresh clone.
+            $backupDir = "$InstallDir.broken-" + (Get-Date -Format "yyyyMMdd-HHmmss")
+            Write-Warn "Existing directory at $InstallDir is not a valid git repo."
+            Write-Warn "Moving it aside to $backupDir before re-cloning."
             try {
-                Remove-Item -Recurse -Force $InstallDir -ErrorAction Stop
+                Move-Item -LiteralPath $InstallDir -Destination $backupDir -ErrorAction Stop
             } catch {
-                Write-Err "Could not remove $InstallDir : $_"
+                Write-Err "Could not move $InstallDir aside : $_"
                 Write-Info "Close any programs that might be using files in $InstallDir (editors,"
                 Write-Info "terminals, running hermes processes) and try again."
                 throw
@@ -1711,7 +1771,7 @@ function Write-BootstrapMarker {
 function Copy-ConfigTemplates {
     Write-Info "Setting up configuration files..."
     
-    # Create ~/.hermes directory structure
+    # Create the HERMES_HOME directory structure ($HermesHome, default %LOCALAPPDATA%\hermes)
     New-Item -ItemType Directory -Force -Path "$HermesHome\cron" | Out-Null
     New-Item -ItemType Directory -Force -Path "$HermesHome\sessions" | Out-Null
     New-Item -ItemType Directory -Force -Path "$HermesHome\logs" | Out-Null
@@ -1729,13 +1789,13 @@ function Copy-ConfigTemplates {
         $examplePath = "$InstallDir\.env.example"
         if (Test-Path $examplePath) {
             Copy-Item $examplePath $envPath
-            Write-Success "Created ~/.hermes/.env from template"
+            Write-Success "Created $envPath from template"
         } else {
             New-Item -ItemType File -Force -Path $envPath | Out-Null
-            Write-Success "Created ~/.hermes/.env"
+            Write-Success "Created $envPath"
         }
     } else {
-        Write-Info "~/.hermes/.env already exists, keeping it"
+        Write-Info "$envPath already exists, keeping it"
     }
     
     # Create config.yaml
@@ -1744,10 +1804,10 @@ function Copy-ConfigTemplates {
         $examplePath = "$InstallDir\cli-config.yaml.example"
         if (Test-Path $examplePath) {
             Copy-Item $examplePath $configPath
-            Write-Success "Created ~/.hermes/config.yaml from template"
+            Write-Success "Created $configPath from template"
         }
     } else {
-        Write-Info "~/.hermes/config.yaml already exists, keeping it"
+        Write-Info "$configPath already exists, keeping it"
     }
     
     # Create SOUL.md if it doesn't exist (global persona file).
@@ -1780,25 +1840,25 @@ Delete the contents (or this file) to use the default personality.
 "@
         $utf8NoBom = New-Object System.Text.UTF8Encoding($false)
         [System.IO.File]::WriteAllText($soulPath, $soulContent, $utf8NoBom)
-        Write-Success "Created ~/.hermes/SOUL.md (edit to customize personality)"
+        Write-Success "Created $soulPath (edit to customize personality)"
     }
     
-    Write-Success "Configuration directory ready: ~/.hermes/"
+    Write-Success "Configuration directory ready: $HermesHome"
     
-    # Seed bundled skills into ~/.hermes/skills/ (manifest-based, one-time per skill)
-    Write-Info "Syncing bundled skills to ~/.hermes/skills/ ..."
+    # Seed bundled skills into $HermesHome\skills (manifest-based, one-time per skill)
+    Write-Info "Syncing bundled skills to $HermesHome\skills ..."
     $pythonExe = "$InstallDir\venv\Scripts\python.exe"
     if (Test-Path $pythonExe) {
         try {
             & $pythonExe "$InstallDir\tools\skills_sync.py" 2>$null
-            Write-Success "Skills synced to ~/.hermes/skills/"
+            Write-Success "Skills synced to $HermesHome\skills"
         } catch {
             # Fallback: simple directory copy
             $bundledSkills = "$InstallDir\skills"
             $userSkills = "$HermesHome\skills"
             if ((Test-Path $bundledSkills) -and -not (Get-ChildItem $userSkills -Exclude '.bundled_manifest' -ErrorAction SilentlyContinue)) {
                 Copy-Item -Path "$bundledSkills\*" -Destination $userSkills -Recurse -Force -ErrorAction SilentlyContinue
-                Write-Success "Skills copied to ~/.hermes/skills/"
+                Write-Success "Skills copied to $HermesHome\skills"
             }
         }
     }
@@ -2234,6 +2294,24 @@ function Install-Desktop {
                 $code = $LASTEXITCODE
             }
         }
+        # Still failing and the user hasn't pinned their own mirror: GitHub's
+        # Electron release host is likely blocked/throttled (the repeating
+        # "retrying" log). Retry once via npmmirror.com — the de-facto Electron
+        # community mirror (Alibaba). @electron/get SHASUM-checks the download,
+        # but the SHASUMS come from the same mirror, so that guards against a
+        # corrupt/partial download, NOT a compromised mirror: an explicit trust
+        # trade-off we only make AFTER the canonical GitHub download has failed,
+        # and we never override a user-pinned ELECTRON_MIRROR.
+        if ($code -ne 0 -and -not $env:ELECTRON_MIRROR) {
+            $prevMirror = $env:ELECTRON_MIRROR
+            $env:ELECTRON_MIRROR = "https://npmmirror.com/mirrors/electron/"
+            Write-Warn "Desktop build still failing - the Electron download from GitHub looks blocked."
+            Write-Warn "Retrying once via a public Electron mirror ($($env:ELECTRON_MIRROR)):"
+            Write-Info "  (set ELECTRON_MIRROR yourself to use a different/trusted mirror)"
+            & $npmExe run pack 2>&1 | ForEach-Object { "$_" } | Tee-Object -FilePath $buildLog
+            $code = $LASTEXITCODE
+            $env:ELECTRON_MIRROR = $prevMirror
+        }
         $ErrorActionPreference = $prevEAP
         if ($code -ne 0) {
             $errText = Get-Content $buildLog -Raw -ErrorAction SilentlyContinue

scripts/install.sh

@@ -1092,6 +1092,18 @@ show_manual_install_hint() {
 clone_repo() {
     log_info "Installing to $INSTALL_DIR..."
 
+    # An interrupted previous clone leaves a .git with no initial commit, where
+    # the update path's `git stash` / `git checkout` abort with "You do not
+    # have the initial commit yet" and fail the install (#40998). Move such a
+    # partial checkout aside -- never delete it, in case it holds something the
+    # user wants -- so the fresh-clone path below can proceed.
+    if [ -d "$INSTALL_DIR/.git" ] && ! git -C "$INSTALL_DIR" rev-parse --verify HEAD >/dev/null 2>&1; then
+        backup_dir="${INSTALL_DIR}.broken-$(date -u +%Y%m%d-%H%M%S)"
+        log_warn "Existing checkout at $INSTALL_DIR has no commits (interrupted clone)."
+        log_warn "Moving it aside to $backup_dir before re-cloning."
+        mv "$INSTALL_DIR" "$backup_dir"
+    fi
+
     if [ -d "$INSTALL_DIR" ]; then
         if [ -d "$INSTALL_DIR/.git" ]; then
             log_info "Existing installation found, updating..."
@@ -1106,7 +1118,12 @@ clone_repo() {
                 autostash_ref="stash@{0}"
             fi
 
-            git fetch origin
+            # Fetch only the target branch. A bare `git fetch origin` pulls
+            # every ref, and this repo carries thousands of auto-generated
+            # branches — on a non-single-branch checkout that turns each update
+            # into a multi-minute download that can stall the installer.
+            git remote set-branches origin "$BRANCH" 2>/dev/null || true
+            git fetch origin "$BRANCH"
             git checkout "$BRANCH"
             git pull --ff-only origin "$BRANCH"
 
@@ -2334,6 +2351,93 @@ postinstall_mode() {
     fi
 }
 
+# Clear the cached Electron download + any half-written unpacked output so the
+# next `npm run pack` re-downloads and re-stages from scratch. A corrupt zip in
+# the per-user Electron download cache - most often a partial/resumed download
+# that leaves concatenated junk - makes electron-builder's `unpack-electron`
+# extract a tree MISSING the electron binary, so the `electron`->`Hermes` rename
+# dies with ENOENT and every re-run repeats the broken extraction forever. This
+# is the bash sibling of install.ps1's Clear-ElectronBuildCache and the Python
+# _purge_electron_build_cache() used by `hermes desktop`; install.sh was the only
+# build path lacking it. Echoes the removed paths (one per line); best-effort.
+clear_electron_build_cache() {
+    local desktop_dir="$1"
+    local removed=""
+
+    # Per-user Electron download cache dirs, honoring the overrides @electron/get
+    # respects, then the platform defaults (macOS: ~/Library/Caches/electron,
+    # Linux: $XDG_CACHE_HOME/electron or ~/.cache/electron).
+    local cache_dirs=()
+    [ -n "${electron_config_cache:-}" ] && cache_dirs+=("$electron_config_cache")
+    [ -n "${ELECTRON_CACHE:-}" ] && cache_dirs+=("$ELECTRON_CACHE")
+    if [ "$OS" = "macos" ]; then
+        cache_dirs+=("$HOME/Library/Caches/electron")
+    else
+        [ -n "${XDG_CACHE_HOME:-}" ] && cache_dirs+=("$XDG_CACHE_HOME/electron")
+        cache_dirs+=("$HOME/.cache/electron")
+    fi
+
+    local dir zip
+    for dir in "${cache_dirs[@]}"; do
+        [ -d "$dir" ] || continue
+        # Recurse: the bad copy may be the top-level zip OR a copy inside an
+        # @electron/get hash subdir.
+        while IFS= read -r zip; do
+            [ -n "$zip" ] || continue
+            if rm -f "$zip" 2>/dev/null; then
+                removed="$removed$zip
+"
+            fi
+        done <<EOF
+$(find "$dir" -type f -name 'electron-*.zip' 2>/dev/null)
+EOF
+    done
+
+    # A half-written unpacked dir from an interrupted prior pack poisons the
+    # rename even after the zip is fixed (mac-arm64-unpacked / linux-unpacked).
+    local release_dir="$desktop_dir/release"
+    if [ -d "$release_dir" ]; then
+        local unpacked
+        while IFS= read -r unpacked; do
+            [ -n "$unpacked" ] || continue
+            if rm -rf "$unpacked" 2>/dev/null; then
+                removed="$removed$unpacked
+"
+            fi
+        done <<EOF
+$(find "$release_dir" -maxdepth 1 -type d -name '*-unpacked' 2>/dev/null)
+EOF
+    fi
+
+    printf '%s' "$removed"
+}
+
+# Run the desktop pack in $1 (the apps/desktop dir). `npm run pack` = tsc +
+# vite build + electron-builder --dir, producing an unpacked app for the
+# current OS. Signing auto-discovery is disabled so electron-builder falls back
+# to an ad-hoc signature instead of grabbing an unrelated Developer ID from the
+# keychain (a real signed/notarized .dmg needs Apple credentials — a separate
+# release concern). Optional $2 = an ELECTRON_MIRROR base URL for this attempt,
+# used as a fallback when the default GitHub release download is blocked.
+_desktop_pack() {
+    local desktop_dir="$1"
+    local mirror="${2:-}"
+    if [ -n "$mirror" ]; then
+        ( cd "$desktop_dir" && ELECTRON_MIRROR="$mirror" CSC_IDENTITY_AUTO_DISCOVERY=false npm run pack )
+    else
+        ( cd "$desktop_dir" && CSC_IDENTITY_AUTO_DISCOVERY=false npm run pack )
+    fi
+}
+
+# Public Electron mirror used as a last-resort fallback when GitHub's release
+# host is blocked/throttled (the repeating "retrying" symptom). npmmirror.com is
+# the de-facto Electron community mirror (Alibaba). @electron/get SHASUM-checks
+# the download, but the SHASUMS come from the same mirror — that guards against a
+# corrupt/partial download, NOT a compromised mirror. Reaching for it is an
+# explicit trust trade-off we only make AFTER the canonical GitHub download has
+# failed, and we never override a user-pinned ELECTRON_MIRROR.
+DESKTOP_ELECTRON_FALLBACK_MIRROR="https://npmmirror.com/mirrors/electron/"
+
 # Build apps/desktop into a launchable native app. Mirrors install.ps1's
 # Install-Desktop: a root-level npm install so the apps/* workspace resolves
 # the desktop's own deps (Electron ~150MB), then `npm run pack`
@@ -2391,18 +2495,53 @@ install_desktop() {
     }
     log_success "Desktop workspace dependencies installed"
 
-    # 2. Build. `npm run pack` = tsc + vite build + electron-builder --dir,
-    #    producing an unpacked app for the current OS. We disable signing
-    #    auto-discovery so electron-builder falls back to an ad-hoc signature
-    #    instead of grabbing an unrelated Developer ID from the keychain; a
-    #    real signed/notarized .dmg needs Apple credentials and is a separate
-    #    release concern.
+    # 2. Build, with up to three escalating attempts so a transient/blocked
+    #    Electron download self-heals instead of failing the whole install:
+    #      a) plain `npm run pack` (downloads Electron from GitHub),
+    #      b) on failure, purge a corrupt cached zip + stale unpacked dir and
+    #         retry (matches install.ps1 / `hermes desktop`),
+    #      c) on still-failing, fall back to a public Electron mirror — this is
+    #         the GitHub-blocked/throttled case (the repeating "retrying" log).
     log_info "Building desktop app (this takes 1-3 minutes)..."
-    ( cd "$desktop_dir" && CSC_IDENTITY_AUTO_DISCOVERY=false npm run pack ) || {
+    local pack_ok=false
+    if _desktop_pack "$desktop_dir"; then
+        pack_ok=true
+    else
+        # (b) Corrupt cached Electron zip is the most common self-healable cause.
+        local purged
+        purged="$(clear_electron_build_cache "$desktop_dir")"
+        if [ -n "$purged" ]; then
+            log_warn "Desktop build failed; cleared cached Electron download and retrying once..."
+            if _desktop_pack "$desktop_dir"; then
+                pack_ok=true
+            fi
+        fi
+    fi
+
+    # (c) Still failing and the user hasn't pinned their own mirror: the GitHub
+    #     release host is likely blocked/throttled. Retry once via a public
+    #     Electron mirror (@electron/get still SHASUM-verifies the download).
+    if [ "$pack_ok" = false ] && [ -z "${ELECTRON_MIRROR:-}" ]; then
+        log_warn "Desktop build still failing — the Electron download from GitHub looks blocked."
+        log_warn "Retrying once via a public Electron mirror ($DESKTOP_ELECTRON_FALLBACK_MIRROR)..."
+        log_warn "  (set ELECTRON_MIRROR yourself to use a different/trusted mirror)"
+        if _desktop_pack "$desktop_dir" "$DESKTOP_ELECTRON_FALLBACK_MIRROR"; then
+            pack_ok=true
+        fi
+    fi
+
+    if [ "$pack_ok" = false ]; then
         log_error "Desktop app build failed"
-        log_info "Run manually: cd $desktop_dir && npm run pack"
+        # If the log shows repeated "retrying" lines fetching the Electron zip,
+        # the binary download is blocked/throttled (firewall, proxy, region) and
+        # the mirror fallback above also couldn't reach a host. Try a mirror you
+        # trust and rebuild (@electron/get honors ELECTRON_MIRROR):
+        log_info "If the log shows Electron download retries, rebuild via a reachable mirror:"
+        log_info "  ELECTRON_MIRROR=<mirror-base-url> \\"
+        log_info "    bash -c 'cd \"$desktop_dir\" && CSC_IDENTITY_AUTO_DISCOVERY=false npm run pack'"
+        log_info "Otherwise build manually: cd $desktop_dir && npm run pack"
         return 1
-    }
+    fi
 
     local app=""
     if [ "$OS" = "linux" ]; then

scripts/release.py

@@ -45,26 +45,45 @@ ACP_REGISTRY_MANIFEST = REPO_ROOT / "acp_registry" / "agent.json"
 
 # Auto-extracted from noreply emails + manual overrides
 AUTHOR_MAP = {
+    "peterhao@Peters-MacBook-Air.local": "pinguarmy",
+    "barronlroth@gmail.com": "barronlroth",
+    "ondrej.drapalik@gmail.com": "OndrejDrapalik",
+    "tomasz.panek@gmail.com": "tomekpanek",
+    "philipadsouza@gmail.com": "PhilipAD",
+    "zhuhaoyu0909@icloud.com": "underthestars-zhy",
+    "raysun12142006@gmail.com": "yanxue06",
+    "alberto.regalado@ymail.com": "ARegalado1",
     "alchemistchaos@protonmail.com": "AlchemistChaos",  # co-author only
     "gilad@smiti.ai": "giladbau",
     "yusufalweshdemir@gmail.com": "Dusk1e",
     "804436395@qq.com": "LaPhilosophie",
     "maxmitcham@mac.home": "maxtrigify",
     "ccook@nvms.com": "ccook1963",
+    "kristian@agrointel.no": "kristianvast",
     "thomas.paquette@gmail.com": "RyTsYdUp",
     "techxacm@gmail.com": "ProgramCaiCai",
     "266365592+bmoore210@users.noreply.github.com": "bmoore210",
+    "123150002+deaneeth@users.noreply.github.com": "deaneeth",
+    "157839748+psionic73@users.noreply.github.com": "psionic73",
     "manishbyatroy@gmail.com": "manishbyatroy",
     "chilltulpa@gmail.com": "TheGardenGallery",
     "al@randomsnowflake.me": "randomsnowflake",
     "zakame@zakame.net": "zakame",
     "152110621+jiangkoumo@users.noreply.github.com": "jiangkoumo",
     "834740219@qq.com": "ViewWay",
+    "matt@vestigial.dev": "m4dni5",
     "harjoth.khara@gmail.com": "harjothkhara",
     "129007007+HeLLGURD@users.noreply.github.com": "HeLLGURD",
     "290859878+synapsesx@users.noreply.github.com": "synapsesx",
     "dirtyren@users.noreply.github.com": "dirtyren",
+    "mvanhorn@MacBook-Pro.local": "mvanhorn",
+    "470766206@qq.com": "youjunxiaji",
+    "mharris@parallel.ai": "NormallyGaussian",
+    "roger@roger.local": "mollusk",
+    "ted.malone@outlook.com": "temalo",
+    "adityamalik2833@gmail.com": "alarcritty",
     "islam666@users.noreply.github.com": "islam666",
+    "mnajafian@nvidia.com": "mnajafian-nv",
     "25539605+lsaether@users.noreply.github.com": "lsaether",
     "30080538+JimStenstrom@users.noreply.github.com": "JimStenstrom",
     "rod.boev@gmail.com": "rodboev",
@@ -95,6 +114,7 @@ AUTHOR_MAP = {
     "metalclaudbot@gmail.com": "HashClawAI",
     "tonybear55665566@gmail.com": "TonyPepeBear",
     "kaspersniels@gmail.com": "nielskaspers",
+    "daxxpasquini@gmail.com": "bpasquini",
     "kurobaryo@gmail.com": "kurobaryo",
     "scubamount@users.noreply.github.com": "scubamount",
     "251514042+youngstar-eth@users.noreply.github.com": "youngstar-eth",
@@ -119,6 +139,7 @@ AUTHOR_MAP = {
     "wangpuv@hotmail.com": "wangpuv",
     "202622897+ticketclosed-wontfix@users.noreply.github.com": "ticketclosed-wontfix",
     "wuxuebin1993@gmail.com": "victorGPT",
+    "xiaoxingitee@gmail.com": "xiaoxinova",
     "wei.chen.coder@gmail.com": "wenchengxucool",
     "frowte3k@gmail.com": "Frowtek",
     "211828103+julio-cloudvisor@users.noreply.github.com": "julio-cloudvisor",
@@ -927,6 +948,8 @@ AUTHOR_MAP = {
     "michel.belleau@malaiwah.com": "malaiwah",
     "gnanasekaran.sekareee@gmail.com": "gnanam1990",
     "jz.pentest@gmail.com": "0xyg3n",
+    "56406949+RaumfahrerSpiffy@users.noreply.github.com": "Spaceman-Spiffy",  # PR #35586 (renamed account)
+    "ian@culling.ca": "ianculling",  # PR #36087
     "7093928+0xyg3n@users.noreply.github.com": "0xyg3n",
     "nftpoetrist@gmail.com": "nftpoetrist",  # PR #18982
     "millerc79@users.noreply.github.com": "millerc79",  # PR #19033
@@ -966,6 +989,7 @@ AUTHOR_MAP = {
     "draixagent@gmail.com": "draix",
     "martin.alca@gmail.com": "draix",
     "junminliu@gmail.com": "JimLiu",
+    "juraj@bednar.io": "jooray",
     "jarvischer@gmail.com": "maxchernin",
     "levantam.98.2324@gmail.com": "LVT382009",
     "zhurongcheng@rcrai.com": "heykb",
@@ -1026,6 +1050,7 @@ AUTHOR_MAP = {
     "zhang9w0v5@qq.com": "zhang9w0v5",
     "fuleinist@outlook.com": "fuleinist",
     "43494187+Llugaes@users.noreply.github.com": "Llugaes",
+    "xiangji.chen@centurygame.com": "Llugaes",
     "fengtianyu88@users.noreply.github.com": "fengtianyu88",
     "l.moncany@gmail.com": "lmoncany",
     "fatinghenji@users.noreply.github.com": "fatinghenji",
@@ -1086,6 +1111,7 @@ AUTHOR_MAP = {
     "holynn@placeholder.local": "holynn-q",
     "agent@hermes.local": "jacdevos",
     "sunsky.lau@gmail.com": "liuhao1024",
+    "rob@rbrtbn.com": "rbrtbn",
     "haaasined@gmail.com": "VinciZhu",
     "fabianoeq@gmail.com": "rodrigoeqnit",
     "178342791+sgtworkman@users.noreply.github.com": "sgtworkman",
@@ -1235,6 +1261,7 @@ AUTHOR_MAP = {
     "charliekerfoot@gmail.com": "CharlieKerfoot",  # PR #18951
     # Debug share upload-time redaction (May 2026)
     "dhuysamen@gmail.com": "GodsBoy",  # PR #19318
+    "github@nadyahermes.anonaddy.com": "ruangraung",  # PR #42308
     "mrcoferland@gmail.com": "mrcoferland",  # PR #19023
     "chenlinfeng@ruije.com.cn": "noOne-list",  # PR #19050
     "briansu@Mac-mini.attlocal.net": "likejudy",  # PR #19052
@@ -1257,6 +1284,7 @@ AUTHOR_MAP = {
     "leon@sgp43.com": "LeonSGP43",  # PR #18739 salvage of #14570
     "miniding@miniding.home": "Foolafroos",  # PR #20329 French locale
     "montbra@gmail.com": "Montbra",  # PR #20897 salvage of #16189 (TUI voice PTT)
+    "275835513+paulb26@users.noreply.github.com": "paulb26",  # PR #24135 salvage (pty-bridge killpg)
     "promptsiren@gmail.com": "firefly",  # PR #18123 salvage of #16660 (ContextVars)
     "wtyopenclaw@gmail.com": "WuTianyi123",  # PR #20275 salvage of #13723 (feishu markdown)
     "zhicheng.han@mathematik.uni-goettingen.de": "hanzckernel",  # PR #20311 (api-server approval events)
@@ -1462,6 +1490,7 @@ AUTHOR_MAP = {
     # v0.15.0 additions
     "glen@workmanfirearms.com": "sgtworkman",
     "jorge.fuenmayort@gmail.com": "jfuenmayor",
+    "josh.dow@prepad.io": "joshuadow",  # PR #43004 salvage (desktop WS session rebind)
     "mordred@inaugust.com": "emonty",
     "rodrigoeq@hotmail.com": "rodrigoeqnit",
     "soliva.johnpaul@icloud.com": "jonpol01",
@@ -1485,6 +1514,10 @@ AUTHOR_MAP = {
     "leonard@sellem.me": "leonardsellem",  # PR #37405 (desktop WS origin guard on remote/Tailscale binds)
     "42903577+ohMyJason@users.noreply.github.com": "ohMyJason",  # PR #29810 (discover_models in custom_providers section 4)
     "singhsanidhya741@gmail.com": "sanidhyasin",  # PR #40403 salvage (model.default_headers for custom OpenAI-compatible providers, #40033)
+    "josephjohnson.joel@gmail.com": "JoelJJohnson",  # PR #39913 salvage (Windows ConPTY dashboard chat bridge)
+    "andreas@schwarz-ketsch.de": "Nea74",  # PR #40022 co-author credit (same Windows ConPTY bridge design)
+    "chanhokyim@gmail.com": "joel611",  # PR #33958 salvage (DISCORD_ALLOWED_ROLES role_authorized gateway flag)
+    "desg38@gmail.com": "dschnurbusch",  # PR #42373 salvage (archive compressed conversation lineages)
 }
 
 

scripts/run_tests_parallel.py

@@ -246,6 +246,98 @@ def _kill_tree(proc: "subprocess.Popen", pgid: int | None = None) -> None:
         pass
 
 
+def _spawn_pytest_once(
+    cmd: List[str],
+    repo_root: Path,
+    file_timeout: float,
+    *,
+    timeout_note: str = "per-file timeout",
+) -> Tuple[int, str]:
+    """Run one ``pytest`` subprocess to completion and return ``(rc, output)``.
+
+    Spawns the child in its own process group / session so a hung file and
+    its grandchildren (uvicorn servers, async runtimes, etc.) can be SIGKILL'd
+    as a tree on timeout rather than orphaning onto PID 1. Shared by the
+    primary per-file run and the exit-4 retry loop so the lifecycle/cleanup
+    logic lives in exactly one place.
+    """
+    proc = subprocess.Popen(
+        cmd,
+        cwd=repo_root,
+        stdout=subprocess.PIPE,
+        stderr=subprocess.STDOUT,
+        text=True,
+        # POSIX: place the child at the head of its own process group so
+        # _kill_tree can SIGKILL the group atomically.
+        # Windows: this maps to CREATE_NEW_PROCESS_GROUP in CPython 3.12+;
+        # _kill_tree handles the Windows path via taskkill /F /T.
+        start_new_session=True,
+    )
+
+    # Capture the pgid NOW, before the leader can exit and be reaped. Once
+    # the leader is reaped, os.getpgid(proc.pid) raises ProcessLookupError
+    # even though grandchildren in that group are still alive — defeating
+    # the whole cleanup. None on Windows where the pgid concept doesn't apply.
+    pgid: int | None = None
+    if sys.platform != "win32":
+        try:
+            pgid = os.getpgid(proc.pid)
+        except (ProcessLookupError, PermissionError):
+            pgid = None
+
+    try:
+        output, _ = proc.communicate(timeout=file_timeout)
+        rc = proc.returncode
+    except subprocess.TimeoutExpired:
+        _kill_tree(proc, pgid=pgid)
+        try:
+            output, _ = proc.communicate(timeout=10)
+        except subprocess.TimeoutExpired:
+            output = "(file timeout exceeded; output unavailable)"
+        rc = 124  # de facto convention for "killed by timeout".
+        output = (
+            f"({timeout_note}: {file_timeout:.0f}s exceeded; "
+            f"process tree SIGKILL'd)\n{output}"
+        )
+    except BaseException:
+        # KeyboardInterrupt / runner crash — make sure no zombie
+        # grandchildren outlive us.
+        _kill_tree(proc, pgid=pgid)
+        raise
+    else:
+        # Happy path: pytest exited on its own. Kill the group anyway in
+        # case it left grandchildren behind; already-dead is a no-op.
+        _kill_tree(proc, pgid=pgid)
+
+    return rc, output
+
+
+# How many times to re-run a file that exits 4 ("file or directory not found")
+# while the file demonstrably exists on disk. On loaded shared CI runners the
+# planner can enumerate a file (tests counted via --collect-only) but the
+# per-file subprocess fail to stat it moments later — and a SINGLE immediate
+# retry can land in the same brief high-load window and fail again. We retry a
+# few times with a short backoff so transient I/O pressure has time to settle.
+_EXIT4_RETRY_ATTEMPTS = 3
+_EXIT4_RETRY_BACKOFF_SECONDS = 0.5
+
+
+def _file_present(file: Path, *, attempts: int = 3, delay: float = 0.2) -> bool:
+    """Return True if ``file`` exists, re-checking a few times.
+
+    ``Path.exists()`` itself issues a ``stat`` that can transiently fail under
+    the same load that makes pytest report "file or directory not found", so a
+    single negative check is not authoritative. Only conclude the file is
+    genuinely missing if it's absent across several spaced checks.
+    """
+    for i in range(attempts):
+        if file.exists():
+            return True
+        if i < attempts - 1:
+            time.sleep(delay)
+    return False
+
+
 def _run_one_file(
     file: Path,
     pytest_args: List[str],
@@ -280,60 +372,60 @@ def _run_one_file(
     """
     cmd = [sys.executable, "-m", "pytest", str(file), *pytest_args]
     subproc_start = time.monotonic()
-    proc = subprocess.Popen(
-        cmd,
-        cwd=repo_root,
-        stdout=subprocess.PIPE,
-        stderr=subprocess.STDOUT,
-        text=True,
-        # POSIX: place the child at the head of its own process group so
-        # _kill_tree can SIGKILL the group atomically.
-        # Windows: this maps to CREATE_NEW_PROCESS_GROUP in CPython 3.12+;
-        # _kill_tree handles the Windows path via taskkill /F /T.
-        start_new_session=True,
-    )
+    rc, output = _spawn_pytest_once(cmd, repo_root, file_timeout)
 
-    # Capture the pgid NOW, before the leader can exit and be reaped.
-    # Once the leader is reaped, os.getpgid(proc.pid) raises
-    # ProcessLookupError even though grandchildren in that group are
-    # still alive — defeating the whole cleanup. None on Windows where
-    # the pgid concept doesn't apply (taskkill walks ppid chain instead).
-    pgid: int | None = None
-    if sys.platform != "win32":
-        try:
-            pgid = os.getpgid(proc.pid)
-        except (ProcessLookupError, PermissionError):
-            # Astonishingly fast child? Already dead. _kill_tree's
-            # fallback will handle this case as a no-op.
-            pgid = None
-
-    try:
-        output, _ = proc.communicate(timeout=file_timeout)
-        rc = proc.returncode
-    except subprocess.TimeoutExpired:
-        _kill_tree(proc, pgid=pgid)
-        # Drain whatever the child wrote before we killed it so we have
-        # something to surface in the failure dump.
-        try:
-            output, _ = proc.communicate(timeout=10)
-        except subprocess.TimeoutExpired:
-            output = "(file timeout exceeded; output unavailable)"
-        rc = 124  # de facto convention for "killed by timeout".
-        output = (
-            f"(per-file timeout: {file_timeout:.0f}s exceeded; "
-            f"process tree SIGKILL'd)\n{output}"
+    # pytest exit 4 = "file or directory not found" at exec time. On loaded
+    # shared CI runners we have seen the planner enumerate a file (its tests
+    # counted via --collect-only) but the per-file subprocess fail to stat it
+    # moments later — a transient the deterministic LPT slicer otherwise
+    # reproduces on every rerun (same file set → same shard). Re-run the file a
+    # few times with a short backoff so the I/O pressure has time to settle,
+    # but ONLY while the file demonstrably exists on disk. A single immediate
+    # retry (the old behaviour) could land in the same brief high-load window
+    # and fail again; a single Path.exists() check could itself be a flaky stat
+    # under that load, so we re-check existence across spaced attempts.
+    # We do NOT widen the exit-5 rule: exit 4 on a file that genuinely does not
+    # exist must still fail.
+    attempt = 0
+    while rc == 4 and attempt < _EXIT4_RETRY_ATTEMPTS and _file_present(file):
+        attempt += 1
+        time.sleep(_EXIT4_RETRY_BACKOFF_SECONDS * attempt)
+        rc, output = _spawn_pytest_once(
+            cmd, repo_root, file_timeout,
+            timeout_note=f"per-file timeout on exit-4 retry {attempt}",
         )
-    except BaseException:
-        # KeyboardInterrupt / runner crash — make sure no zombie
-        # grandchildren outlive us.
-        _kill_tree(proc, pgid=pgid)
-        raise
-    else:
-        # Happy path: pytest exited on its own. The child process already
-        # cleaned up its grandchildren if it's well-behaved, but
-        # well-behaved is not universal — kill the group anyway. Already-
-        # dead processes are a no-op.
-        _kill_tree(proc, pgid=pgid)
+
+    if rc == 4:
+        # Exit-4 survived the retries (or the file was judged absent).
+        # Capture filesystem forensics so a CI-only "file not found" can
+        # be diagnosed from the log instead of guessed at: does the file
+        # exist NOW, what does the parent dir hold, and is the git tree
+        # clean?  (June 2026: a PR-added test file repeatedly hit exit 4
+        # on one CI shard while passing locally — these lines exist so
+        # the next occurrence is attributable.)
+        forensics = [f"--- exit-4 forensics for {file} ---"]
+        try:
+            forensics.append(f"exists={file.exists()} retries_used={attempt}")
+            parent = file.parent
+            if parent.exists():
+                names = sorted(p.name for p in parent.iterdir())
+                sibling_hint = [n for n in names if file.stem[:12] in n]
+                forensics.append(
+                    f"parent={parent} entries={len(names)} "
+                    f"similar={sibling_hint[:5]}"
+                )
+            else:
+                forensics.append(f"parent={parent} MISSING")
+            git_st = subprocess.run(
+                ["git", "status", "--porcelain"],
+                cwd=repo_root, capture_output=True, text=True, timeout=10,
+            )
+            dirty = git_st.stdout.strip().splitlines()
+            forensics.append(f"git_dirty_entries={len(dirty)}")
+            forensics.extend(f"  {line}" for line in dirty[:10])
+        except Exception as exc:  # noqa: BLE001 — forensics must never mask rc=4
+            forensics.append(f"(forensics error: {exc})")
+        output = output + "\n" + "\n".join(forensics)
 
     if rc == 5:
         # No tests collected — every test in the file was filtered out.

scripts/whatsapp-bridge/bridge.js

@@ -24,7 +24,8 @@ import { Boom } from '@hapi/boom';
 import pino from 'pino';
 import path from 'path';
 import { mkdirSync, readFileSync, writeFileSync, existsSync, readdirSync, unlinkSync } from 'fs';
-import { randomBytes } from 'crypto';
+import { fileURLToPath } from 'url';
+import { randomBytes, createHash } from 'crypto';
 import { execSync } from 'child_process';
 import { tmpdir } from 'os';
 import qrcode from 'qrcode-terminal';
@@ -45,9 +46,28 @@ const WHATSAPP_DEBUG =
 
 const PORT = parseInt(getArg('port', '3000'), 10);
 const SESSION_DIR = getArg('session', path.join(process.env.HOME || '~', '.hermes', 'whatsapp', 'session'));
-const IMAGE_CACHE_DIR = path.join(process.env.HOME || '~', '.hermes', 'image_cache');
-const DOCUMENT_CACHE_DIR = path.join(process.env.HOME || '~', '.hermes', 'document_cache');
-const AUDIO_CACHE_DIR = path.join(process.env.HOME || '~', '.hermes', 'audio_cache');
+// Cache directories: the Python gateway passes the profile-aware paths via
+// env (HERMES_HOME-aware, new cache/ layout).  Fall back to the legacy
+// hardcoded locations for bridges launched outside the gateway.
+const IMAGE_CACHE_DIR = process.env.HERMES_IMAGE_CACHE_DIR
+  || path.join(process.env.HOME || '~', '.hermes', 'image_cache');
+const DOCUMENT_CACHE_DIR = process.env.HERMES_DOCUMENT_CACHE_DIR
+  || path.join(process.env.HOME || '~', '.hermes', 'document_cache');
+const AUDIO_CACHE_DIR = process.env.HERMES_AUDIO_CACHE_DIR
+  || path.join(process.env.HOME || '~', '.hermes', 'audio_cache');
+
+// Self-hash of this script file.  Reported in /health so the Python gateway
+// can detect a running bridge that predates the current bridge.js and
+// restart it instead of silently reusing stale code (stale-bridge trap:
+// `hermes update` updates bridge.js on disk but a long-lived bridge process
+// keeps serving the old behavior forever).
+let SCRIPT_HASH = '';
+try {
+  SCRIPT_HASH = createHash('sha256')
+    .update(readFileSync(fileURLToPath(import.meta.url)))
+    .digest('hex')
+    .slice(0, 16);
+} catch {}
 const PAIR_ONLY = args.includes('--pair-only');
 const WHATSAPP_MODE = getArg('mode', process.env.WHATSAPP_MODE || 'self-chat'); // "bot" or "self-chat"
 const ALLOWED_USERS = parseAllowedUsers(process.env.WHATSAPP_ALLOWED_USERS || '');
@@ -700,6 +720,7 @@ app.get('/health', (req, res) => {
     status: connectionState,
     queueLength: messageQueue.length,
     uptime: process.uptime(),
+    scriptHash: SCRIPT_HASH,
   });
 });