merge next into fix/1240-fleet-transport-check

Resolves conflicts from #1229 (tools/install.sh node provisioning) and #1252
(package.json test:framework-shell). tools/install.sh resolved to the reviewed
composite blob 5d28f773; package.json resolved as a union so both #1252's four
suites and #1245's transport-check suite run (48 links).

packages/mosaic/__tests__/platform/detect.test.ts

@@ -0,0 +1,74 @@
+import { describe, it, expect, vi, beforeEach, afterEach } from 'vitest';
+
+// homedir/platform are read at call time, so they can be stubbed per case.
+vi.mock('node:os', async (importOriginal) => {
+  const actual = await importOriginal<typeof import('node:os')>();
+  return {
+    ...actual,
+    homedir: () => '/home/tester',
+    platform: () => mockPlatform,
+  };
+});
+
+let mockPlatform: NodeJS.Platform = 'linux';
+
+const { getShellProfilePath, detectShell } = await import('../../src/platform/detect.js');
+
+describe('getShellProfilePath', () => {
+  const originalShell = process.env['SHELL'];
+  const originalZdotdir = process.env['ZDOTDIR'];
+
+  beforeEach(() => {
+    mockPlatform = 'linux';
+    delete process.env['ZDOTDIR'];
+  });
+
+  afterEach(() => {
+    if (originalShell === undefined) delete process.env['SHELL'];
+    else process.env['SHELL'] = originalShell;
+    if (originalZdotdir === undefined) delete process.env['ZDOTDIR'];
+    else process.env['ZDOTDIR'] = originalZdotdir;
+  });
+
+  // The regression this guards: setupPath() in stages/finalize.ts appends the
+  // PATH export to whatever this returns. A line written to ~/.bashrc is
+  // unreachable to `bash -lc`, systemd units and agent seats, because Debian's
+  // default .bashrc returns early for non-interactive shells — so an install
+  // reported success and left `mosaic: command not found`. Same for .zshrc,
+  // which zsh only reads for interactive shells.
+  it('never targets an interactive-only rc file', () => {
+    for (const shell of ['/bin/bash', '/usr/bin/zsh']) {
+      process.env['SHELL'] = shell;
+      const profile = getShellProfilePath();
+      expect(profile).not.toMatch(/\.bashrc$/);
+      expect(profile).not.toMatch(/\.zshrc$/);
+    }
+  });
+
+  it('uses ~/.profile for bash', () => {
+    process.env['SHELL'] = '/bin/bash';
+    expect(getShellProfilePath()).toBe('/home/tester/.profile');
+  });
+
+  it('uses ~/.zshenv for zsh', () => {
+    process.env['SHELL'] = '/usr/bin/zsh';
+    expect(getShellProfilePath()).toBe('/home/tester/.zshenv');
+  });
+
+  it('honours ZDOTDIR for zsh', () => {
+    process.env['SHELL'] = '/usr/bin/zsh';
+    process.env['ZDOTDIR'] = '/custom/zdot';
+    expect(getShellProfilePath()).toBe('/custom/zdot/.zshenv');
+  });
+
+  it('falls back to ~/.profile for an unknown shell', () => {
+    process.env['SHELL'] = '/bin/somethingelse';
+    expect(detectShell()).toBe('unknown');
+    expect(getShellProfilePath()).toBe('/home/tester/.profile');
+  });
+
+  it('still routes fish to its own config', () => {
+    process.env['SHELL'] = '/usr/bin/fish';
+    expect(getShellProfilePath()).toBe('/home/tester/.config/fish/config.fish');
+  });
+});

packages/mosaic/framework/install.sh

@@ -35,6 +35,18 @@ SOURCE_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
 TARGET_DIR="${MOSAIC_HOME:-$HOME/.config/mosaic}"
 INSTALL_MODE="${MOSAIC_INSTALL_MODE:-prompt}"
 
+# Normalize the ambient umask so directory modes are a property of the installer
+# and not of whatever shell invoked it (#1236). Debian/Ubuntu ship umask 002, so
+# every `mkdir -p` below yielded 0775 — and the fleet env boundary rejects any
+# managed directory with `mode & 0o022`, which made `mosaic fleet init --write`
+# impossible on a stock install of those distros. Fedora/RHEL ship 022 and did
+# not trip it, so the product worked or did not depending on the operator's
+# login shell. 022 is what this script already assumes it produces: see the
+# umask note in make_durable_snapshot, which restores to the ambient value
+# precisely so "every later sync copy and new framework dir" gets 0644/0755.
+# Now that value is 022 rather than whatever was inherited.
+umask 022
+
 # Deliberately parsed from "$@" (a real, explicit, per-invocation argument) —
 # never an environment variable — so this opt-out can never sit silently
 # inherited in a shell profile. See #869 Point-1 C2.
@@ -696,6 +708,52 @@ sync_framework
 mkdir -p "$TARGET_DIR/memory"
 mkdir -p "$TARGET_DIR/credentials"
 
+# Three directories must be 0700, not merely not-group-writable (#1236).
+# The fleet code guards them with two different masks in two different
+# languages, and the strict one wins:
+#
+#   assertPrivateManagedDirectory (fleet-reconciler.js, `mode & 0o077`)
+#       -> MOSAIC_HOME and MOSAIC_HOME/fleet, checked before the roster lock is
+#          taken, so every mutating `mosaic fleet` command dies at 0755.
+#   assert_private_directory (tools/fleet/start-agent-session.sh, `mode & 077`)
+#       -> MOSAIC_HOME/fleet/agents, checked before a pane is ever spawned.
+#
+# Their laxer siblings (`mode & 0o022`) accept 0755, which is why normalizing
+# the umask above is necessary and not sufficient — a correct umask-022 install
+# still produces 0755 and still cannot run `mosaic fleet init --write`. Say the
+# strict modes outright rather than inferring them from a umask.
+#
+# Only these. The rest of the tree is content, stays 0755, and is only ever
+# reached by the 0o022 checks, which 0755 satisfies.
+chmod 700 "$TARGET_DIR" 2>/dev/null || \
+  warn "Could not set 0700 on $TARGET_DIR — 'mosaic fleet' mutations will fail as unsafe-permissions."
+if [[ -d "$TARGET_DIR/fleet" ]]; then
+  chmod 700 "$TARGET_DIR/fleet" 2>/dev/null || \
+    warn "Could not set 0700 on $TARGET_DIR/fleet — 'mosaic fleet' mutations will fail as unsafe-permissions."
+fi
+# fleet/agents does not exist on a first install — the CLI creates it 0700 on
+# demand. It is chmod'd here for the UPGRADE case: a tree built under umask 002
+# has it at 0775, and the repair sweep below cannot rescue it, because stripping
+# group/other write from 0755 leaves 0750 and `mode & 077` is still non-zero.
+if [[ -d "$TARGET_DIR/fleet/agents" ]]; then
+  chmod 700 "$TARGET_DIR/fleet/agents" 2>/dev/null || \
+    warn "Could not set 0700 on $TARGET_DIR/fleet/agents — agent sessions will fail to start as unsafe-permissions."
+fi
+# credentials/ holds secrets and was never meant to be group-readable either.
+# It is not on the fleet boundary, so a failure here breaks nothing — but it is
+# the one directory where a silently-failed chmod leaves secrets group-readable,
+# which is precisely the failure worth a line in the output.
+chmod 700 "$TARGET_DIR/credentials" 2>/dev/null || \
+  warn "Could not set 0700 on $TARGET_DIR/credentials — stored secrets may be readable by other users on this host."
+
+# Repair an existing tree. The umask above only governs directories this run
+# creates, so a host installed under umask 002 before this fix keeps its 0775
+# dirs through every upgrade and stays broken. Strips group/other WRITE only —
+# never read or execute — so it can repair the boundary violation without
+# changing who can traverse or read anything. Scoped to directories: file modes
+# are the manifest's business, not this fix's.
+find "$TARGET_DIR" -type d -perm /022 -exec chmod go-w {} + 2>/dev/null || true
+
 # Reconcile contract files from defaults/ into the framework root: framework-owned
 # files (CONSTITUTION/AGENTS/STANDARDS) are overwritten every upgrade (a divergent
 # copy is backed up once); user-seeded files (TOOLS) are written on first install only.

packages/mosaic/framework/systemd/user/mosaic-agent@.service

@@ -4,6 +4,14 @@ Documentation=https://git.mosaicstack.dev/mosaicstack/stack
 Requires=mosaic-tmux-holder.service
 After=mosaic-tmux-holder.service
 PartOf=mosaic-tmux-holder.service
+# Do not attempt a seat before its generated env exists. `install` enables this
+# unit (WantedBy=default.target) but on a roster-v2 fleet the reconciler owns the
+# generated env, so between `install` and the first `apply`/`regen --write` there
+# is a boot window where ExecStart would run against an absent env file and the
+# launcher would fail the unit. A skipped unit is the honest state for "enabled
+# but not yet configured"; systemd re-evaluates the condition on every start, so
+# the seat comes up on the next start once the reconciler has written env.
+ConditionPathExists=%h/.config/mosaic/fleet/agents/%i.env.generated
 
 [Service]
 Type=oneshot

packages/mosaic/framework/systemd/user/test-fleet-units.sh

@@ -128,6 +128,14 @@ EOF
 sleep 30
 EOF
   chmod 700 "$AGENT_BIN/mosaic"
+  # The launcher resolves the roster's runtime against PANE_PATH before it
+  # spawns anything (#1241), so the runtime this projection names has to be
+  # present here even though the fake `mosaic` above never execs it.
+  cat > "$AGENT_BIN/pi" <<'EOF'
+#!/bin/sh
+sleep 30
+EOF
+  chmod 700 "$AGENT_BIN/pi"
   server_environment_before=$(tmux -L "$TEST_SOCKET" show-environment -g | sort)
   server_sessions_before=$(tmux -L "$TEST_SOCKET" list-sessions | sort)
   if /usr/bin/env -i HOME="$HOLDER_HOME" PATH=/usr/bin:/bin MOSAIC_HOME="$AGENT_HOME" \

packages/mosaic/framework/tools/fleet/start-agent-session.sh

@@ -286,6 +286,36 @@ _build_runtime_bin_prefix() {
 MOSAIC_RUNTIME_BIN_PREFIX=$(_build_runtime_bin_prefix)
 PANE_PATH=${MOSAIC_RUNTIME_BIN_PREFIX:+${MOSAIC_RUNTIME_BIN_PREFIX}:}/usr/local/bin:/usr/bin:/bin
 
+# #1241. The pane runs `mosaic yolo <runtime>` under PANE_PATH with a cleared
+# environment. A binary missing from *that* path is a pane that dies in under a
+# second, inside a session nobody is attached to, with its diagnostic scrolled
+# into a pane tmux then destroys. Resolve both here, before any effect, where
+# the failure is still attributable to the thing that caused it.
+#
+# `mosaic yolo <runtime>` runs checkRuntime(runtime) and the binary it looks for
+# is named exactly like the runtime, so resolving the runtime name is the same
+# question the pane will ask a moment later — asked while an operator can still
+# see the answer.
+_resolve_in_pane_path() {
+  PATH="$PANE_PATH" command -v -- "$1" 2>/dev/null
+}
+
+# Exit 69 (EX_UNAVAILABLE): the seat cannot be provided. Distinguished from the
+# 64 (EX_USAGE) rejections above, which mean the projection itself was bad —
+# here the data is fine and the host is not ready. Callers tell the individual
+# cases apart by `code=`, the same way fail_env's many codes share exit 64.
+fail_launch() {
+  local code="$1"
+  shift
+  echo "ERROR: agent launch aborted: code=${code} agent=${AGENT_NAME} $*" >&2
+  exit 69
+}
+
+for required_binary in mosaic "$MOSAIC_AGENT_RUNTIME"; do
+  _resolve_in_pane_path "$required_binary" >/dev/null ||
+    fail_launch missing-binary "'${required_binary}' is not on the pane PATH (${PANE_PATH})"
+done
+
 _ensure_claude_workdir_trusted() {
   local workdir="$1"
   local resolved
@@ -384,6 +414,19 @@ if [ -n "$PANE_PID" ]; then
   _start_heartbeat_sidecar "$AGENT_NAME" "$PANE_PID" \
     "$MOSAIC_HEARTBEAT_RUN_DIR" "$MOSAIC_HEARTBEAT_INTERVAL" || \
     echo "WARNING: heartbeat sidecar could not be started for $AGENT_NAME" >&2
+elif _tmux has-session -t "=${AGENT_NAME}:0.0" 2>/dev/null; then
+  # #1241. Session present, no pane PID after a second of retries. Whatever this
+  # is, it is not a seat an operator can use, so it is not a success either.
+  fail_launch pane-pid-unresolved \
+    "tmux reports the session but no pane PID after 5 attempts"
 else
-  echo "WARNING: could not resolve pane PID for $AGENT_NAME — heartbeat sidecar not started" >&2
+  # #1241. This branch used to print a WARNING about the heartbeat sidecar and
+  # exit 0. It is not a heartbeat problem: tmux destroys a session when its pane
+  # command exits, so an absent session one second after new-session means the
+  # runtime died on startup. Reporting it as success is what let `fleet start`
+  # return 0 over three dead panes — the launcher knew, and said the wrong thing
+  # at the wrong severity to the wrong layer.
+  fail_launch pane-did-not-survive \
+    "the pane exited immediately and tmux destroyed the session;" \
+    "run 'mosaic yolo ${MOSAIC_AGENT_RUNTIME}' in ${MOSAIC_AGENT_WORKDIR} to see why"
 fi

packages/mosaic/framework/tools/fleet/test-start-agent-session.sh

@@ -23,8 +23,26 @@ index=0
 if [ "${args[0]:-}" = -L ]; then index=2; fi
 case "${args[$index]:-}" in
   has-session)
+    # The holder always answers. MOSAIC_TEST_HELD_SESSIONS lets a case add
+    # other targets that should answer too — without it there is no way to
+    # model "tmux still reports the session" for a non-holder agent, and the
+    # launcher's pane-pid-unresolved branch is unreachable from this harness.
+    #
+    # A listed target answers only AFTER new-session, because the launcher asks
+    # this question twice about the same name: once before launching, where a
+    # yes means "already running, nothing to do, exit 0", and once after, where
+    # a yes means "the session survived". A shim that answered yes to both
+    # would short-circuit at the first and never reach the branch under test —
+    # it would look like coverage and measure the idempotency path instead.
     for argument in "${args[@]}"; do
       [ "$argument" = '=_holder:0.0' ] && exit 0
+      case " ${MOSAIC_TEST_HELD_SESSIONS:-} " in
+        *" $argument "*)
+          if tr '\0' '\n' < "${MOSAIC_TEST_TMUX_CALLS:?}" | grep -qxF new-session; then
+            exit 0
+          fi
+          ;;
+      esac
     done
     exit 1
     ;;
@@ -62,6 +80,30 @@ env -0 > "${MOSAIC_HOME:?}/fleet/pane-environment"
 SHIM
 chmod +x "$FAKE_BIN/mosaic"
 
+# The runtime the rosters below name. The launcher resolves it against PANE_PATH
+# before spawning (#1241), so it has to exist somewhere the pane would find it —
+# not merely on the launcher's own PATH.
+printf '#!/usr/bin/env bash\nexit 0\n' > "$FAKE_BIN/pi"
+chmod +x "$FAKE_BIN/pi"
+
+# PANE_PATH is derived partly from `npm config get prefix`. Left to the real npm
+# it would splice whatever the host has installed into the path under test, and
+# the missing-binary cases below would pass or fail by accident of the machine.
+cat > "$FAKE_BIN/npm" <<'SHIM'
+#!/usr/bin/env bash
+printf '%s\n' "${MOSAIC_TEST_NPM_PREFIX:-/nonexistent}"
+SHIM
+chmod +x "$FAKE_BIN/npm"
+
+# PANE_PATH always ends in the system path. A host that installs these there can
+# not measure the missing-binary cases at all, and a green run would mean
+# nothing — so say so instead of passing.
+for host_binary in mosaic pi; do
+  if PATH=/usr/local/bin:/usr/bin:/bin command -v "$host_binary" >/dev/null 2>&1; then
+    fail "host provides '$host_binary' in the system path; missing-binary cases are not measurable here"
+  fi
+done
+
 write_generated() {
   local home="$1"
   local agent="$2"
@@ -81,6 +123,19 @@ MOSAIC_TMUX_SOCKET=mosaic-test
 EOF
   chmod 600 "$home/fleet/agents/$agent.env.generated"
   mkdir -p "$home/work"
+  install_pane_binaries "$home"
+}
+
+# `$PANE_HOME/.npm-global/bin` is one of the prefixes the launcher folds into
+# PANE_PATH, so this is the pane's own view of "installed", distinct from the
+# launcher's PATH. Tests that need a binary *absent* remove it from here.
+install_pane_binaries() {
+  local pane_home="$1"
+  mkdir -p "$pane_home/.npm-global/bin"
+  local binary
+  for binary in mosaic pi; do
+    ln -sf "$FAKE_BIN/$binary" "$pane_home/.npm-global/bin/$binary"
+  done
 }
 
 run_start() {
@@ -88,6 +143,7 @@ run_start() {
   local agent="$2"
   HOME="$home" PATH="$FAKE_BIN:$PATH" MOSAIC_TEST_TMUX_CALLS="$TMUX_CALLS" \
     MOSAIC_TEST_PANE_PID="${MOSAIC_TEST_PANE_PID:-}" \
+    MOSAIC_TEST_HELD_SESSIONS="${MOSAIC_TEST_HELD_SESSIONS:-}" \
     MOSAIC_TEST_HOME="$home" \
     MOSAIC_TEST_FLEET_OWNER=123e4567-e89b-12d3-a456-426614174000 \
     MOSAIC_HOME="$home" "$START" "$agent"
@@ -98,7 +154,10 @@ run_start() {
 HOME_VALID="$ROOT/valid"
 AGENT_VALID="coder0"
 write_generated "$HOME_VALID" "$AGENT_VALID"
-run_start "$HOME_VALID" "$AGENT_VALID"
+# A live pane PID is part of what "valid launch" means. Until #1241 this case
+# ran with none, so the suite's one success path was itself a dead pane the
+# launcher reported as fine.
+MOSAIC_TEST_PANE_PID=$$ run_start "$HOME_VALID" "$AGENT_VALID"
 valid_args=$(tr '\0' '\n' < "$TMUX_CALLS")
 echo "$valid_args" | grep -qF new-session || fail "valid generated projection did not reach tmux"
 echo "$valid_args" | grep -qF 'mosaic' || fail "fixed mosaic launcher command missing"
@@ -245,6 +304,13 @@ PANE_BASH_ENV="$ROOT/pane-boundary.bash-env"
 printf 'MOSAIC_RUNTIME_BIN=%s\n' "$FAKE_BIN" > \
   "$HOME_PANE_BOUNDARY/fleet/agents/coder-pane-boundary.env.local"
 chmod 600 "$HOME_PANE_BOUNDARY/fleet/agents/coder-pane-boundary.env.local"
+# This case does not go through run_start, so its pane binaries come from
+# MOSAIC_RUNTIME_BIN=$FAKE_BIN in the env.local written above — not from the
+# symlinks install_pane_binaries planted under the generated home, which this
+# launcher never consults because HOME here is the trusted parent. That is a
+# legitimate resolution path, but it means dropping MOSAIC_RUNTIME_BIN from
+# this case on the belief that the symlinks cover it would break the #1241
+# binary check rather than exercise it.
 LD_PRELOAD='/not/loaded/by-clean-bootstrap.so' \
 BASH_ENV="$PANE_BASH_ENV" \
 MOSAIC_UNTRUSTED_SENTINEL='must-not-reach-pane' \
@@ -258,6 +324,7 @@ PATH="$PANE_STALE_PATH" \
     "MOSAIC_TEST_HOME=$PANE_TRUSTED_HOME" \
     MOSAIC_TEST_FLEET_OWNER=123e4567-e89b-12d3-a456-426614174000 \
     MOSAIC_TEST_EXECUTE_PANE=1 \
+    "MOSAIC_TEST_PANE_PID=$$" \
     "$START" coder-pane-boundary
 pane_args=$(tr '\0' '\n' < "$TMUX_CALLS")
 echo "$pane_args" | grep -qxF "HOME=$PANE_TRUSTED_HOME" || \
@@ -392,6 +459,75 @@ echo "$interaction_policy_args" | grep -qF 'new-session' && \
 echo "$output" | grep -qF 'operator interaction service requires runtime pi' || \
   fail "interaction pinned-policy check did not follow strict parsing"
 
+# #1241. The pane runs `mosaic yolo <runtime>` against PANE_PATH. A binary
+# missing from that path is a launch failure, and it has to be named before the
+# session is created — after it, the diagnostic dies with the pane.
+assert_missing_pane_binary_rejected() {
+  local binary="$1"
+  local home="$ROOT/missing-$binary"
+  local agent="coder-missing-$binary"
+  write_generated "$home" "$agent"
+  rm -f "$home/.npm-global/bin/$binary"
+
+  : > "$TMUX_CALLS"
+  local output
+  if output=$(MOSAIC_TEST_PANE_PID=$$ run_start "$home" "$agent" 2>&1); then
+    fail "launch succeeded with '$binary' absent from the pane PATH"
+  fi
+  echo "$output" | grep -qF 'code=missing-binary' || fail "missing '$binary' diagnostic missing"
+  echo "$output" | grep -qF "'$binary'" || fail "missing-binary diagnostic did not name $binary"
+  if tr '\0' '\n' < "$TMUX_CALLS" | grep -qF new-session; then
+    fail "launcher created a session it knew would die ($binary absent)"
+  fi
+}
+
+assert_missing_pane_binary_rejected mosaic
+assert_missing_pane_binary_rejected pi
+
+# #1241. tmux destroys a session when its pane command exits, so no pane PID a
+# second after new-session means the runtime died on startup. This used to be a
+# WARNING about the heartbeat sidecar followed by exit 0 — three layers above it
+# then reported a fleet that was not running.
+: > "$TMUX_CALLS"
+HOME_DEAD_PANE="$ROOT/dead-pane"
+write_generated "$HOME_DEAD_PANE" "coder-dead-pane"
+if output=$(MOSAIC_TEST_PANE_PID='' run_start "$HOME_DEAD_PANE" coder-dead-pane 2>&1); then
+  fail "launcher reported success over a pane that did not survive"
+fi
+echo "$output" | grep -qF 'code=pane-did-not-survive' || fail "dead-pane diagnostic missing"
+if echo "$output" | grep -qiF 'heartbeat'; then
+  fail "dead pane is still being reported as a heartbeat-sidecar problem"
+fi
+tr '\0' '\n' < "$TMUX_CALLS" | grep -qF new-session || \
+  fail "dead-pane case did not reach the launch it is measuring"
+
+# #1241, the other way a pane fails. Above, tmux destroyed the session and
+# has-session said so. Here the session is still there and no PID comes back
+# after the retries — a different fault (the pane is alive but unusable, or
+# tmux is answering inconsistently) that an operator has to be told apart from
+# a runtime that died on startup.
+#
+# This case exists because the branch that handles it shipped with nothing able
+# to reach it: the shim answered has-session only for the holder, so every
+# non-holder agent landed in the session-is-gone branch no matter what. A
+# defensive branch nothing exercises is the same shape as the bug this whole
+# change is about, one layer down.
+: > "$TMUX_CALLS"
+HOME_NO_PID="$ROOT/pane-no-pid"
+write_generated "$HOME_NO_PID" "coder-no-pid"
+if output=$(MOSAIC_TEST_PANE_PID='' MOSAIC_TEST_HELD_SESSIONS='=coder-no-pid:0.0' \
+  run_start "$HOME_NO_PID" coder-no-pid 2>&1); then
+  fail "launcher reported success over a session with no resolvable pane PID"
+fi
+echo "$output" | grep -qF 'code=pane-pid-unresolved' || \
+  fail "session-present/no-PID was not reported as pane-pid-unresolved: $output"
+if echo "$output" | grep -qF 'code=pane-did-not-survive'; then
+  fail "a session tmux still reports was diagnosed as a destroyed session"
+fi
+if echo "$output" | grep -qiF 'heartbeat'; then
+  fail "an unresolvable pane PID is still being reported as a heartbeat-sidecar problem"
+fi
+
 # Exact stop derives the socket exclusively from the validated generated
 # projection and ignores an ambient socket supplied by the caller.
 : > "$TMUX_CALLS"

packages/mosaic/framework/tools/quality/test-enumeration-exclusions.txt

@@ -32,10 +32,6 @@ packages/mosaic/framework/tools/tmux/test-send-message-socket.sh | requires a re
 packages/mosaic/framework/tools/tmux/test-send-message-verdict.sh | requires real tmux-pane fixtures on a throwaway socket; CI image ships no tmux; #1017 burndown (same condition as its sibling)
 
 # --- single-suite directories: unmeasured in CI ---
-packages/mosaic/framework/tools/fleet/test-start-agent-session.sh | unmeasured in CI image; stubs tmux via a fake bin dir, likely CI-fit; #1017 burndown
-packages/mosaic/framework/tools/glpi/test-list-http-status.sh | unmeasured in CI image; stub-based (#807 regression harness), likely CI-fit; #1017 burndown
-packages/mosaic/framework/tools/orchestrator/test-board-roll.sh | unmeasured in CI image; file-fixture based, likely CI-fit; #1017 burndown
-packages/mosaic/framework/tools/woodpecker/test-ci-wait-exit-matrix.sh | unmeasured in CI image; drives ci-wait.sh against a stub pipeline-status.sh, likely CI-fit; #1017 burndown
 
 # --- naming-boundary files the strict test-*.sh prefix cannot even name ---
 # (#1017: three independent censuses handled the microtest file three different

packages/mosaic/package.json

@@ -25,7 +25,7 @@
     "lint": "eslint src",
     "typecheck": "tsc --noEmit",
     "test": "vitest run --passWithNoTests && pnpm run test:framework-shell",
-    "test:framework-shell": "bash framework/tools/quality/scripts/check-test-enumeration.sh && bash framework/tools/quality/scripts/test-check-test-enumeration.sh && python3 src/lease-broker/daemon_deadline_unittest.py && python3 src/lease-broker/normative_fragments_unittest.py && python3 src/lease-broker/promotion_binding_unittest.py && python3 src/lease-broker/promotion_trigger_unittest.py && python3 src/lease-broker/receipt_challenge_unittest.py && python3 src/lease-broker/context_recovery_unittest.py && python3 src/lease-broker/recovery_runtime_unittest.py && python3 src/lease-broker/recovery_b1_adversarial_unittest.py && python3 src/lease-broker/receipt_observer_client_unittest.py && python3 src/lease-broker/invariant_r_unittest.py && python3 src/lease-broker/framework_skill_portability_unittest.py && python3 src/mutator-gate/runtime_tools_unittest.py && python3 src/mutator-gate/runtime_launch_guard_unittest.py && python3 src/mutator-gate/version_coupling_unittest.py && python3 framework/tools/lease-broker/check-runtime-launches.py --root ../.. && bash framework/tools/codex/test-pr-diff-context.sh && bash framework/tools/qa/test-deps-preflight.sh && bash framework/tools/git/test-pr-review-gitea-comment.sh && bash framework/tools/git/test-pr-review-repo-host-override.sh && bash framework/tools/git/test-ci-queue-wait-branch-absent.sh && bash framework/tools/git/test-ci-queue-wait-tristate.sh && bash framework/tools/git/test-ci-queue-wait-github-checks.sh && bash framework/tools/git/test-pr-merge-queue-branch.sh && bash framework/tools/git/test-pr-merge-head-pin.sh && bash framework/tools/git/test-pr-merge-message-field.sh && bash framework/tools/git/test-git-credential-mosaic.sh && bash framework/tools/git/test-gitea-token-identity.sh && bash framework/tools/woodpecker/test-terminal-green-contract.sh && bash framework/tools/_scripts/test-install-ordering-guard.sh && bash framework/tools/_scripts/test-mosaic-init-rce.sh && bash framework/tools/_scripts/test-fleet-transport-check.sh && bash framework/tools/tmux/agent-send.test.sh && bash framework/tools/wake/test-wake-store-ack.sh && bash framework/tools/wake/test-wake-store-enqueue-race.sh && bash framework/tools/wake/test-wake-digest-hmac.sh && bash framework/tools/wake/test-wake-digest-quarantine.sh && bash framework/tools/wake/test-wake-detector.sh && bash framework/tools/wake/test-wake-fn-oracle.sh && bash framework/tools/wake/test-wake-reconcile.sh && bash framework/tools/wake/test-wake-beacon.sh && bash framework/tools/wake/test-wake-preimage.sh && bash framework/tools/wake/test-wake-install.sh"
+    "test:framework-shell": "bash framework/tools/quality/scripts/check-test-enumeration.sh && bash framework/tools/quality/scripts/test-check-test-enumeration.sh && python3 src/lease-broker/daemon_deadline_unittest.py && python3 src/lease-broker/normative_fragments_unittest.py && python3 src/lease-broker/promotion_binding_unittest.py && python3 src/lease-broker/promotion_trigger_unittest.py && python3 src/lease-broker/receipt_challenge_unittest.py && python3 src/lease-broker/context_recovery_unittest.py && python3 src/lease-broker/recovery_runtime_unittest.py && python3 src/lease-broker/recovery_b1_adversarial_unittest.py && python3 src/lease-broker/receipt_observer_client_unittest.py && python3 src/lease-broker/invariant_r_unittest.py && python3 src/lease-broker/framework_skill_portability_unittest.py && python3 src/mutator-gate/runtime_tools_unittest.py && python3 src/mutator-gate/runtime_launch_guard_unittest.py && python3 src/mutator-gate/version_coupling_unittest.py && python3 framework/tools/lease-broker/check-runtime-launches.py --root ../.. && bash framework/tools/codex/test-pr-diff-context.sh && bash framework/tools/qa/test-deps-preflight.sh && bash framework/tools/git/test-pr-review-gitea-comment.sh && bash framework/tools/git/test-pr-review-repo-host-override.sh && bash framework/tools/git/test-ci-queue-wait-branch-absent.sh && bash framework/tools/git/test-ci-queue-wait-tristate.sh && bash framework/tools/git/test-ci-queue-wait-github-checks.sh && bash framework/tools/git/test-pr-merge-queue-branch.sh && bash framework/tools/git/test-pr-merge-head-pin.sh && bash framework/tools/git/test-pr-merge-message-field.sh && bash framework/tools/git/test-git-credential-mosaic.sh && bash framework/tools/git/test-gitea-token-identity.sh && bash framework/tools/woodpecker/test-terminal-green-contract.sh && bash framework/tools/_scripts/test-install-ordering-guard.sh && bash framework/tools/_scripts/test-mosaic-init-rce.sh && bash framework/tools/tmux/agent-send.test.sh && bash framework/tools/wake/test-wake-store-ack.sh && bash framework/tools/wake/test-wake-store-enqueue-race.sh && bash framework/tools/wake/test-wake-digest-hmac.sh && bash framework/tools/wake/test-wake-digest-quarantine.sh && bash framework/tools/wake/test-wake-detector.sh && bash framework/tools/wake/test-wake-fn-oracle.sh && bash framework/tools/wake/test-wake-reconcile.sh && bash framework/tools/wake/test-wake-beacon.sh && bash framework/tools/wake/test-wake-preimage.sh && bash framework/tools/wake/test-wake-install.sh && bash framework/tools/fleet/test-start-agent-session.sh && bash framework/tools/glpi/test-list-http-status.sh && bash framework/tools/orchestrator/test-board-roll.sh && bash framework/tools/woodpecker/test-ci-wait-exit-matrix.sh && bash framework/tools/_scripts/test-fleet-transport-check.sh"
   },
   "dependencies": {
     "@mosaicstack/brain": "workspace:*",

packages/mosaic/src/commands/fleet-roster-v2-dispatch.spec.ts

@@ -0,0 +1,323 @@
+import { execFile } from 'node:child_process';
+import { mkdir, mkdtemp, readFile, readdir, rm, stat, writeFile } from 'node:fs/promises';
+import { tmpdir } from 'node:os';
+import { join, resolve } from 'node:path';
+import { Command } from 'commander';
+import { afterEach, describe, expect, it, vi } from 'vitest';
+import { registerFleetCommand, type CommandResult, type CommandRunner } from './fleet.js';
+
+/**
+ * #1237: the v1-only commands (`ps`, `install`, `install-systemd`, `add`,
+ * `remove`) rejected a roster-v2 fleet outright, so a greenfield v2 box could
+ * never get its units placed. These tests pin the three behaviours that fix
+ * gives it, and the two it deliberately does NOT give it.
+ *
+ * The load-bearing negative is that `install` on v2 writes no generated env:
+ * the reconciler owns that file through projectRosterV2AgentGeneratedEnv, and a
+ * second writer here — necessarily through the v1 mapping — is exactly the
+ * drift the #791 single-SSOT invariant exists to prevent.
+ */
+
+const rosterV2 = `
+version: 2
+generation: 4
+transport: tmux
+tmux:
+  socket_name: mosaic-fleet
+  holder_session: _holder
+defaults:
+  working_directory: /srv/mosaic
+  runtime: pi
+runtimes:
+  pi:
+    reset_command: /new
+agents:
+  - name: coder0
+    alias: Coder 0
+    class: code
+    runtime: pi
+    provider: openai
+    model: gpt-5.6-sol
+    reasoning: high
+    tool_policy: code
+    working_directory: /srv/mosaic
+    persistent_persona: false
+    reset_between_tasks: true
+    lifecycle:
+      enabled: true
+      desired_state: stopped
+    launch:
+      yolo: true
+  - name: coder1
+    alias: Coder 1
+    class: code
+    runtime: pi
+    provider: openai
+    model: gpt-5.6-sol
+    reasoning: medium
+    tool_policy: code
+    working_directory: /srv/other
+    persistent_persona: false
+    reset_between_tasks: true
+    lifecycle:
+      enabled: true
+      desired_state: stopped
+    launch:
+      yolo: true
+`;
+
+let tempHome: string | undefined;
+const savedHome = process.env.HOME;
+const savedMosaicHome = process.env.MOSAIC_HOME;
+
+afterEach(async (): Promise<void> => {
+  vi.restoreAllMocks();
+  process.exitCode = undefined;
+  if (savedHome === undefined) delete process.env.HOME;
+  else process.env.HOME = savedHome;
+  if (savedMosaicHome === undefined) delete process.env.MOSAIC_HOME;
+  else process.env.MOSAIC_HOME = savedMosaicHome;
+  if (tempHome) await rm(tempHome, { recursive: true, force: true });
+  tempHome = undefined;
+});
+
+/**
+ * A HOME with a roster-v2 fleet and nothing else — the greenfield shape, before
+ * anything has been installed, applied or started.
+ */
+async function v2Home(): Promise<string> {
+  tempHome = await mkdtemp(join(tmpdir(), 'mosaic-fleet-v2-dispatch-'));
+  process.env.HOME = tempHome;
+  delete process.env.MOSAIC_HOME;
+  const mosaicHome = join(tempHome, '.config', 'mosaic');
+  for (const directory of ['fleet', 'fleet/agents', 'fleet/roles']) {
+    await mkdir(join(mosaicHome, directory), { recursive: true, mode: 0o700 });
+  }
+  await writeFile(join(mosaicHome, 'fleet', 'roster.yaml'), rosterV2, { mode: 0o600 });
+  await writeFile(join(mosaicHome, 'fleet', 'roles', 'code.md'), '`class: code`\n\n# code\n', {
+    mode: 0o600,
+  });
+  return mosaicHome;
+}
+
+/**
+ * Stands in for a box where nothing is running: every systemctl and tmux probe
+ * fails the way it does before the holder has ever started. `ps` must survive
+ * this — it is the command an operator reaches for to find out *why* there is
+ * no seat, so it has to report the emptiness rather than fail on it.
+ */
+const greenfieldRunner: CommandRunner = async (command): Promise<CommandResult> => {
+  if (command === 'tmux') {
+    return { stdout: '', stderr: 'no server running on /tmp/tmux-1000/mosaic-fleet', exitCode: 1 };
+  }
+  return { stdout: '', stderr: '', exitCode: 1 };
+};
+
+function program(runner: CommandRunner = greenfieldRunner): Command {
+  const result = new Command();
+  result.exitOverride();
+  registerFleetCommand(result, { runner, frameworkRoot: resolve(process.cwd(), 'framework') });
+  return result;
+}
+
+function capture(): string[] {
+  const lines: string[] = [];
+  vi.spyOn(console, 'log').mockImplementation((value: string): void => {
+    lines.push(value);
+  });
+  return lines;
+}
+
+async function exists(path: string): Promise<boolean> {
+  try {
+    await stat(path);
+    return true;
+  } catch {
+    return false;
+  }
+}
+
+describe('mosaic fleet ps — roster v2', (): void => {
+  it('lists every v2 agent on a greenfield box with nothing running, and does not throw', async (): Promise<void> => {
+    await v2Home();
+    const lines = capture();
+
+    await expect(
+      program().parseAsync(['node', 'mosaic', 'fleet', 'ps', '--json']),
+    ).resolves.toBeDefined();
+
+    const rows = JSON.parse(lines.join('\n')) as {
+      name: string;
+      runtime: string;
+      alias?: string;
+      paneAlive: boolean;
+      source: string;
+    }[];
+    expect(rows.map((row) => row.name).sort()).toEqual(['coder0', 'coder1']);
+    // The v2 roster's per-agent fields must survive the read model, not be
+    // flattened into defaults.
+    expect(rows.every((row) => row.runtime === 'pi')).toBe(true);
+    expect(rows.find((row) => row.name === 'coder0')?.alias).toBe('Coder 0');
+    // Nothing is running, and that is a report, not an error.
+    expect(rows.every((row) => row.paneAlive === false)).toBe(true);
+    expect(rows.every((row) => row.source === 'roster')).toBe(true);
+    expect(process.exitCode ?? 0).toBe(0);
+  });
+});
+
+describe('mosaic fleet install — roster v2', (): void => {
+  it('places the tool files and unit templates', async (): Promise<void> => {
+    const mosaicHome = await v2Home();
+    capture();
+
+    await expect(
+      program().parseAsync(['node', 'mosaic', 'fleet', 'install', '--no-enable']),
+    ).resolves.toBeDefined();
+
+    // Units live in the systemd user dir, not under the Mosaic home.
+    const systemdUserDir = join(tempHome!, '.config', 'systemd', 'user');
+    for (const unit of [
+      'mosaic-tmux-holder.service',
+      '[email protected]',
+      '[email protected]',
+    ]) {
+      expect(await exists(join(systemdUserDir, unit))).toBe(true);
+    }
+    const launcher = join(mosaicHome, 'tools', 'fleet', 'start-agent-session.sh');
+    expect(await exists(launcher)).toBe(true);
+    expect((await stat(launcher)).mode & 0o777).toBe(0o755);
+  });
+
+  it('writes NO generated env — that file belongs to the reconciler (#791)', async (): Promise<void> => {
+    const mosaicHome = await v2Home();
+    capture();
+
+    await program().parseAsync(['node', 'mosaic', 'fleet', 'install', '--no-enable']);
+
+    const agentDir = join(mosaicHome, 'fleet', 'agents');
+    expect(await readdir(agentDir)).toEqual([]);
+  });
+
+  it('tells the operator which command does own the env', async (): Promise<void> => {
+    await v2Home();
+    const lines = capture();
+
+    await program().parseAsync(['node', 'mosaic', 'fleet', 'install', '--no-enable']);
+
+    expect(lines.join('\n')).toContain('mosaic fleet apply');
+  });
+});
+
+describe('[email protected]', (): void => {
+  const unitPath = resolve(process.cwd(), 'framework', 'systemd', 'user', '[email protected]');
+
+  /** The single `ConditionPathExists=` value declared by the unit template. */
+  async function conditionPath(): Promise<string> {
+    const unit = await readFile(unitPath, 'utf8');
+    const matches = unit.match(/^ConditionPathExists=(.+)$/gm) ?? [];
+    expect(matches).toHaveLength(1);
+    return matches[0]!.slice('ConditionPathExists='.length).trim();
+  }
+
+  it('will not attempt a seat before the reconciler has written its env', async (): Promise<void> => {
+    // The pairing that makes "install writes no env" safe: install enables the
+    // unit (WantedBy=default.target) but does not start it, so without this
+    // condition a reboot between `install` and the first `apply` would run
+    // ExecStart against an absent env file and fail every seat unit.
+    expect(await conditionPath()).toBe('%h/.config/mosaic/fleet/agents/%i.env.generated');
+  });
+
+  /**
+   * The two halves of the guard's *effect*, which no assertion on the literal
+   * string can cover on its own.
+   *
+   * Measured end to end on a real box (canary, 2026-08-16) rather than inferred:
+   * with the condition, `systemctl --user start mosaic-agent@<name>` on an agent
+   * with no generated env returns rc=0, `Result=success`, `ConditionResult=no`,
+   * and journals "skipped, unmet condition check". With the condition removed by
+   * drop-in and nothing else changed, the same start returns rc=1,
+   * `Result=exit-code`, `ExecMainStatus=64`, and the unit enters `failed`.
+   *
+   * systemd is not available in this suite, so these two tests pin the parts
+   * that can drift in code: the condition naming a *different* file than the one
+   * the fleet actually writes, and the launcher quietly becoming tolerant of an
+   * absent env — either of which turns the condition into decoration while the
+   * literal-string assertion above still passes.
+   */
+  it('guards exactly the file the fleet writes, so the two cannot drift apart', async (): Promise<void> => {
+    const mosaicHome = await v2Home();
+    const rendered = (await conditionPath()).replace('%h', tempHome!).replace('%i', 'coder0');
+
+    // The path an installed fleet actually places for this agent.
+    expect(rendered).toBe(join(mosaicHome, 'fleet', 'agents', 'coder0.env.generated'));
+  });
+
+  it('guards a real failure — the launcher rejects an absent generated env', async (): Promise<void> => {
+    await v2Home();
+    await program().parseAsync(['node', 'mosaic', 'fleet', 'install', '--no-enable']);
+
+    // Exactly what ExecStart runs, against the state the condition exists to
+    // catch: unit enabled, reconciler has not written env yet.
+    const launched = await new Promise<{ code: number | null; stderr: string }>((settle) => {
+      const child = execFile(
+        '/bin/bash',
+        [
+          '--noprofile',
+          '--norc',
+          join(tempHome!, '.config', 'mosaic', 'tools', 'fleet', 'start-agent-session.sh'),
+          'coder0',
+        ],
+        { env: { HOME: tempHome!, MOSAIC_AGENT_NAME: 'coder0', PATH: '/usr/bin:/bin' } },
+        (_error, _stdout, stderr) => {
+          settle({ code: child.exitCode, stderr });
+        },
+      );
+    });
+
+    expect(launched.code).not.toBe(0);
+    expect(launched.stderr).toContain('missing-file');
+  });
+});
+
+describe('mosaic fleet add / remove — roster v2', (): void => {
+  it('add refuses, and names the two-step v2 sequence instead of inventing defaults', async (): Promise<void> => {
+    await v2Home();
+
+    await expect(
+      program().parseAsync([
+        'node',
+        'mosaic',
+        'fleet',
+        'add',
+        'coder2',
+        '--runtime',
+        'pi',
+        '--class',
+        'code',
+      ]),
+    ).rejects.toThrow(/mosaic fleet create[\s\S]*mosaic fleet apply/);
+  });
+
+  it('remove refuses, and names delete plus apply', async (): Promise<void> => {
+    await v2Home();
+
+    await expect(
+      program().parseAsync(['node', 'mosaic', 'fleet', 'remove', 'coder1']),
+    ).rejects.toThrow(/mosaic fleet delete coder1[\s\S]*mosaic fleet apply/);
+  });
+
+  // Note: this one passes on the unmodified tree too — there `remove` throws in
+  // the v1 parser, before it can touch anything. It is a regression guard on the
+  // ordering of the new guard clause, not evidence that the fix works.
+  it('refuses BEFORE mutating the roster', async (): Promise<void> => {
+    const mosaicHome = await v2Home();
+    const rosterPath = join(mosaicHome, 'fleet', 'roster.yaml');
+    const before = await readFile(rosterPath, 'utf8');
+
+    await expect(
+      program().parseAsync(['node', 'mosaic', 'fleet', 'remove', 'coder1']),
+    ).rejects.toThrow();
+
+    expect(await readFile(rosterPath, 'utf8')).toBe(before);
+  });
+});

packages/mosaic/src/commands/fleet.ts

@@ -34,6 +34,7 @@ export {
   resolveInstalledFleetRosterPath,
 } from '../fleet/fleet-roster-v1.js';
 export type { FleetAgent, FleetRoster } from '../fleet/fleet-roster-v1.js';
+import { parseRosterV2 } from '../fleet/roster-v2.js';
 import {
   registerFleetAgentCrudCommands,
   type FleetAgentCrudCommandDeps,
@@ -820,7 +821,7 @@ export function buildEnableLingerCommand(user: string): string[] {
  */
 export async function enableFleetUnits(
   runner: CommandRunner,
-  roster: FleetRoster,
+  roster: { readonly agents: readonly { readonly name: string }[] },
   opts: { enable?: boolean },
 ): Promise<void> {
   if (opts.enable === false) {
@@ -1527,7 +1528,8 @@ export function registerFleetCommand(program: Command, deps: FleetCommandDeps =
     .option('--no-enable', 'Skip enabling units for boot-survival')
     .action(async (opts: { enable?: boolean }) => {
       await installFleet(cmd, frameworkRoot);
-      const roster = await loadRosterForCommand(cmd);
+      // Unit enablement needs agent names only, so it reads either version.
+      const roster = await loadRosterReadModel(cmd);
       await enableFleetUnits(runner, roster, opts);
     });
 
@@ -1537,7 +1539,8 @@ export function registerFleetCommand(program: Command, deps: FleetCommandDeps =
     .option('--no-enable', 'Skip enabling units for boot-survival')
     .action(async (opts: { enable?: boolean }) => {
       await installFleet(cmd, frameworkRoot);
-      const roster = await loadRosterForCommand(cmd);
+      // Unit enablement needs agent names only, so it reads either version.
+      const roster = await loadRosterReadModel(cmd);
       await enableFleetUnits(runner, roster, opts);
     });
 
@@ -1688,7 +1691,9 @@ export function registerFleetCommand(program: Command, deps: FleetCommandDeps =
     .action(async (opts: { json?: boolean }) => {
       const commandOpts = cmd.opts<{ mosaicHome: string; roster?: string }>();
       const activePaths = resolveFleetPaths(commandOpts.mosaicHome);
-      const roster = await loadRosterForCommand(cmd);
+      // ps only reads, so it takes the version-agnostic read model rather than
+      // the v1 parser, which rejects a v2 roster outright.
+      const roster = await loadRosterReadModel(cmd);
       const { tenant_id, host } = getDefaultTenantAndHost();
       const nowMs = Date.now();
 
@@ -1908,6 +1913,16 @@ export function registerFleetCommand(program: Command, deps: FleetCommandDeps =
           start: boolean;
         },
       ) => {
+        if (await usesRosterV2ControlPlane(cmd)) {
+          // command.error, not a bare throw: this is operator guidance, and a
+          // bare throw reaches the top level uncaught and prints it under a Node
+          // stack trace. Measured on canary — the message is the whole point of
+          // the refusal, so it has to arrive readable.
+          cmd.error(rosterV2MutationGuidance('add', 'create', name), {
+            code: 'fleet.roster-v2',
+            exitCode: 1,
+          });
+        }
         if (!VALID_FLEET_RUNTIMES.includes(opts.runtime)) {
           throw new Error(
             `Invalid runtime "${opts.runtime}". Valid runtimes: ${VALID_FLEET_RUNTIMES.join(', ')}.`,
@@ -1973,6 +1988,12 @@ export function registerFleetCommand(program: Command, deps: FleetCommandDeps =
     .description('Remove an agent from the fleet roster')
     .option('--keep-files', 'Skip deleting env and heartbeat files')
     .action(async (name: string, opts: { keepFiles?: boolean }) => {
+      if (await usesRosterV2ControlPlane(cmd)) {
+        cmd.error(rosterV2MutationGuidance('remove', 'delete', name), {
+          code: 'fleet.roster-v2',
+          exitCode: 1,
+        });
+      }
       const commandOpts = cmd.opts<{ mosaicHome: string; roster?: string }>();
       const activePaths = resolveFleetPaths(commandOpts.mosaicHome);
       const rosterPath = await resolveRosterPath(commandOpts.mosaicHome, commandOpts.roster);
@@ -2331,7 +2352,9 @@ export function registerFleetAgentCommands(
 async function installFleet(cmd: Command, frameworkRoot: string): Promise<void> {
   const activePaths = resolveFleetPaths(cmd.opts<{ mosaicHome: string }>().mosaicHome);
   assertDefaultMosaicHomeForSystemd(activePaths.mosaicHome);
-  const roster = await loadRosterForCommand(cmd);
+  // Read model first: every file this function places is roster-independent, and
+  // the v1 parser would reject a v2 roster before any of them were written.
+  const roster = await loadRosterReadModel(cmd);
   await ensureFleetHolderIdentity(activePaths.mosaicHome);
   await mkdir(activePaths.fleetToolsDir, { recursive: true });
   await mkdir(activePaths.tmuxToolsDir, { recursive: true });
@@ -2391,16 +2414,30 @@ async function installFleet(cmd: Command, frameworkRoot: string): Promise<void>
     join(activePaths.systemdUserDir, '[email protected]'),
   );
 
-  for (const agent of roster.agents) {
+  // On roster v2 the reconciler owns the generated env: `apply` writes it and
+  // `regen` rebuilds it, both from projectRosterV2AgentGeneratedEnv. Writing it
+  // here too — necessarily through the v1 mapping — would be the third writer of
+  // one file and would break the #791 single-SSOT invariant. So v2 gets the tool
+  // files and the units, and nothing else.
+  if (roster.version === 2) {
+    console.log(
+      `Installed fleet tools and systemd units for ${roster.agents.length} agent(s). ` +
+        `Generated env is owned by the reconciler on roster v2 — run: mosaic fleet apply --expected-generation <n>`,
+    );
+    return;
+  }
+
+  const v1Roster = await loadRosterForCommand(cmd);
+  for (const agent of v1Roster.agents) {
     await writeAgentEnvironmentProjection({
       mosaicHome: activePaths.mosaicHome,
       agentEnvDir: activePaths.agentEnvDir,
       agentName: agent.name,
-      generated: generateAgentEnvValues(roster, agent),
+      generated: generateAgentEnvValues(v1Roster, agent),
     });
   }
 
-  console.log(`Installed fleet files for ${roster.agents.length} agent(s).`);
+  console.log(`Installed fleet files for ${v1Roster.agents.length} agent(s).`);
 }
 
 async function loadRosterForCommand(cmd: Command): Promise<FleetRoster> {
@@ -2427,6 +2464,77 @@ async function usesRosterV2ControlPlane(cmd: Command): Promise<boolean> {
   );
 }
 
+/**
+ * `add`/`remove` and `create`/`delete` are not two spellings of one operation.
+ * The v1 pair edits the roster *and* drives systemd; the v2 pair is documented
+ * as changing desired state "without runtime actions", leaving convergence to
+ * `apply`. `add` also collects four fields where a v2 agent requires eleven, so
+ * routing it to `create` would mean inventing provider, alias, reasoning and
+ * tool-policy defaults on the operator's behalf. Refusing with the real command
+ * is honest; silently guessing an agent's provider is not.
+ */
+function rosterV2MutationGuidance(
+  v1Command: 'add' | 'remove',
+  v2Command: 'create' | 'delete',
+  name: string,
+): string {
+  const target = v2Command === 'delete' ? ` ${name}` : '';
+  return (
+    `mosaic fleet ${v1Command} does not operate on a roster-v2 fleet. ` +
+    `Roster v2 separates desired state from convergence:\n` +
+    `  1. mosaic fleet ${v2Command}${target} --expected-generation <current> ` +
+    `${v2Command === 'create' ? "--agent '<json>' " : ''}` +
+    `(edits the roster only)\n` +
+    `  2. mosaic fleet apply --expected-generation <new> (converges systemd and tmux)\n` +
+    `Read the current generation with: mosaic fleet status`
+  );
+}
+
+/**
+ * The read-only fields shared by roster v1 and v2, for the commands that only
+ * ever *read* the roster (`ps`, and unit enablement inside `install`).
+ *
+ * This is deliberately NOT a v2→v1 downshift. A downshifted `FleetRoster` would
+ * be accepted by `generateAgentEnvValues`, and that would make a third writer of
+ * `fleet/agents/<name>.env.generated` — through the v1 mapping — breaking the
+ * #791 single-SSOT invariant that {@link projectRosterV2AgentGeneratedEnv} is
+ * documented to hold. Keeping the read model this small makes that misuse
+ * impossible: there is nothing here to write a roster or an env file back from.
+ */
+interface FleetRosterReadModel {
+  readonly version: 1 | 2;
+  readonly tmux: { readonly socketName: string; readonly holderSession: string };
+  readonly agents: readonly {
+    readonly name: string;
+    readonly alias?: string;
+    readonly runtime: string;
+  }[];
+}
+
+/** Reads either roster version into the shared read-only view. */
+async function loadRosterReadModel(cmd: Command): Promise<FleetRosterReadModel> {
+  const opts = cmd.opts<{ mosaicHome: string; roster?: string }>();
+  const path = await resolveRosterPath(opts.mosaicHome, opts.roster);
+  if (!(await usesRosterV2ControlPlane(cmd))) {
+    const v1 = await loadRosterAtPath(cmd, path);
+    return {
+      version: 1,
+      tmux: { socketName: v1.tmux.socketName, holderSession: v1.tmux.holderSession },
+      agents: v1.agents,
+    };
+  }
+  try {
+    const v2 = parseRosterV2(await readFleetRosterText(path), 'yaml');
+    return {
+      version: 2,
+      tmux: { socketName: v2.tmux.socketName, holderSession: v2.tmux.holderSession },
+      agents: v2.agents,
+    };
+  } catch (error) {
+    reportFleetRosterConfigurationError(cmd, error);
+  }
+}
+
 async function loadRosterFromAgentCommand(
   command: Command,
   mosaicHomeOverride?: string,

packages/mosaic/src/platform/detect.ts

@@ -1,4 +1,3 @@
-import { existsSync } from 'node:fs';
 import { join } from 'node:path';
 import { homedir, platform } from 'node:os';
 
@@ -22,15 +21,18 @@ export function getShellProfilePath(): string | null {
 
   const shell = detectShell();
   switch (shell) {
+    // Both of these deliberately avoid the interactive-only rc files.
+    // Debian's default .bashrc returns early for non-interactive shells, so a
+    // PATH line appended to it never runs for `bash -lc`, systemd units, or
+    // agent seats — an install could report success and still leave `mosaic`
+    // unreachable. .profile is read by login shells and sources .bashrc for
+    // interactive ones, so one line covers both; .zshenv is zsh's equivalent.
     case 'zsh': {
       const zdotdir = process.env['ZDOTDIR'] ?? home;
-      return join(zdotdir, '.zshrc');
+      return join(zdotdir, '.zshenv');
     }
-    case 'bash': {
+    case 'bash':
-      const bashrc = join(home, '.bashrc');
-      if (existsSync(bashrc)) return bashrc;
       return join(home, '.profile');
-    }
     case 'fish':
       return join(home, '.config', 'fish', 'config.fish');
     default:

tools/install.sh

@@ -309,6 +309,87 @@ require_cmd() {
   fi
 }
 
+# True if any shell rc file already puts $1 on PATH.
+#
+# Each file is tested for existence first and grepped one at a time, rather than
+# handed to a single `grep -qs ... "${rc_files[@]}"`. Handing grep a missing file
+# makes the exit status implementation-defined: GNU grep 3.11 returns 0 when -q
+# matched an earlier file, ugrep 7.5 returns 2 for the missing one regardless.
+# On the 2 path the caller reads "not present yet" and appends a duplicate PATH
+# line on every single install.
+path_entry_exists() {
+  local dir="$1" rc_file
+  for rc_file in "$HOME/.profile" "$HOME/.zshenv" "$HOME/.zshrc" "$HOME/.bashrc"; do
+    if [[ -f "$rc_file" ]] && grep -qF "$dir" "$rc_file"; then
+      return 0
+    fi
+  done
+  return 1
+}
+
+# Append `export PATH="$1:$PATH"` to the shell profile so $1 survives this
+# process. An `export` here reaches only the installer; every directory the
+# install leaves behind has to be written down somewhere a later shell reads.
+#
+# Deliberately NOT ~/.bashrc: Debian's default .bashrc returns early for
+# non-interactive shells, so a PATH line appended to the bottom of it is
+# unreachable to `bash -lc`, to systemd units, and to every agent seat — the
+# exact consumers that need these binaries. ~/.profile is read by login shells
+# and Debian's .profile sources .bashrc for interactive ones, so a single line
+# there reaches both. For zsh the always-sourced file is .zshenv, not .zshrc.
+#
+# $1 = directory to add, $2 = label for the comment line.
+# Returns 1 (having warned) if the profile could not be written.
+persist_on_path() {
+  local dir="$1" label="$2" profile
+
+  if path_entry_exists "$dir"; then
+    return 0
+  fi
+
+  if [[ -n "${ZSH_VERSION:-}" ]] || [[ "$(basename "${SHELL:-}")" == "zsh" ]]; then
+    profile="$HOME/.zshenv"
+  else
+    profile="$HOME/.profile"
+  fi
+
+  # Probe writability in a subshell. A redirection failure on a special built-in
+  # aborts the shell it runs in, so it has to be a child; and the redirection on
+  # the subshell is what silences the "Permission denied" the shell would
+  # otherwise print ahead of our own message.
+  if ! ( : >>"$profile" ) 2>/dev/null; then
+    warn "$dir is not on your PATH and $profile could not be written"
+    dim "  Add to your shell rc:  export PATH=\"$dir:\$PATH\""
+    return 1
+  fi
+
+  {
+    echo ""
+    echo "# $label"
+    echo "export PATH=\"$dir:\$PATH\""
+  } >>"$profile"
+  ok "Added $dir to PATH in $profile"
+  return 0
+}
+
+# Persist $PREFIX/bin on PATH instead of only warning about it.
+#
+# The warning it replaces was the last step of an otherwise successful install,
+# so the installer reported success and left `mosaic: command not found` — an
+# unattended install had no operator to read the advice and act on it.
+ensure_prefix_on_path() {
+  if [[ ":$PATH:" == *":$PREFIX/bin:"* ]]; then
+    return
+  fi
+
+  if path_entry_exists "$PREFIX/bin"; then
+    warn "$PREFIX/bin is in your shell profile but not in this shell"
+  elif ! persist_on_path "$PREFIX/bin" "Mosaic CLI"; then
+    return
+  fi
+  dim "  Run:  export PATH=\"$PREFIX/bin:\$PATH\"   (or start a new login shell)"
+}
+
 # Fleet transport binary (#1240).
 #
 # `mosaic fleet --help` reads "Manage the local Mosaic tmux fleet" and every
@@ -553,8 +634,175 @@ install_next_cli_from_registry() {
   ok "Installed @next packages: CLI ${installed_cli}, gateway ${installed_gateway}"
 }
 
+# ─── node bootstrap ───────────────────────────────────────────────────────────
+#
+# Nothing on a greenfield host installs Node.js, yet this installer and the CLI
+# it installs both hard-require it. Measured on a clean Debian 13 image: the
+# installer stopped at `require_cmd node` with "Required command not found" and
+# nothing was installed, with no hint of how to proceed.
+#
+# Inlined rather than factored into a sibling file on purpose: this script is
+# fetched standalone by curl and has nothing to source.
+#
+# No-op when a suitable node is already on PATH, so it never fights an
+# operator's nvm/fnm/distro node.
+
+NODE_ROOT="${MOSAIC_NODE_ROOT:-$HOME/.mosaic/node}"
+NODE_BOOTSTRAP_VERSION="${MOSAIC_NODE_VERSION:-v22.23.2}"
+NODE_MIN_MAJOR="${MOSAIC_NODE_MIN_MAJOR:-20}"
+NODE_DIST_BASE="${MOSAIC_NODE_DIST_BASE:-https://nodejs.org/dist}"
+
+# Major version of the node at $1, or empty if it will not run.
+node_major_of() {
+  local candidate="$1" version
+  version="$("$candidate" -e 'process.stdout.write(process.versions.node)' 2>/dev/null)" || return 0
+  printf '%s' "${version%%.*}"
+}
+
+node_is_suitable() {
+  local major
+  major="$(node_major_of "$1")"
+  [[ -n "$major" ]] && [[ "$major" -ge "$NODE_MIN_MAJOR" ]]
+}
+
+install_node() {
+  local node_os node_arch tarball release_url work_dir extracted target node_bin
+
+  case "$(uname -s)" in
+    Linux)  node_os="linux" ;;
+    Darwin) node_os="darwin" ;;
+    *) fail "Unsupported OS '$(uname -s)'. Install Node.js >= $NODE_MIN_MAJOR manually."; return 1 ;;
+  esac
+
+  # Linux here means glibc. Node's official linux-x64 build is dynamically
+  # linked against glibc, so on musl (Alpine) the binary will not exec — but it
+  # fails visibly: node_is_suitable rejects it and ensure_node exits with
+  # "install Node.js manually". No silent breakage, just a wasted download.
+  # A musl host needs the unofficial build, which is out of scope here.
+  case "$(uname -m)" in
+    x86_64|amd64)  node_arch="x64" ;;
+    aarch64|arm64) node_arch="arm64" ;;
+    armv7l)        node_arch="armv7l" ;;
+    *) fail "Unsupported architecture '$(uname -m)'. Install Node.js >= $NODE_MIN_MAJOR manually."; return 1 ;;
+  esac
+
+  # .tar.gz rather than the smaller .tar.xz: gzip is universally present, xz is
+  # not, and a minimal image is exactly the case this exists to handle.
+  tarball="node-${NODE_BOOTSTRAP_VERSION}-${node_os}-${node_arch}.tar.gz"
+  release_url="${NODE_DIST_BASE}/${NODE_BOOTSTRAP_VERSION}"
+
+  work_dir="$(mktemp -d "${TMPDIR:-/tmp}/mosaic-node-XXXXXX")"
+
+  info "Installing Node.js $NODE_BOOTSTRAP_VERSION ($node_os-$node_arch) to $NODE_ROOT…"
+
+  if ! curl -fsSL "${release_url}/${tarball}" -o "$work_dir/$tarball"; then
+    fail "Download failed: ${release_url}/${tarball}"
+    rm -rf "$work_dir"; return 1
+  fi
+
+  # Trust assumption, stated so nobody has to infer it: this verifies INTEGRITY
+  # (the tarball matches the manifest), not AUTHENTICITY (the manifest is
+  # genuinely Node's). The only thing establishing that is TLS to
+  # $NODE_DIST_BASE. Node publishes SHASUMS256.txt.sig signed by its release
+  # keys and we do not check it, which is on par with nvm but means pointing
+  # MOSAIC_NODE_DIST_BASE at an untrusted mirror has no signature backstop.
+  # Tracked as a hardening follow-up (raised by scooby in the #1229 review).
+  if ! curl -fsSL "${release_url}/SHASUMS256.txt" -o "$work_dir/SHASUMS256.txt"; then
+    fail "Could not fetch SHASUMS256.txt; refusing to install an unverified runtime."
+    rm -rf "$work_dir"; return 1
+  fi
+
+  # Keep only our artifact's line, so a missing entry is an error not a pass.
+  if ! grep " ${tarball}\$" "$work_dir/SHASUMS256.txt" >"$work_dir/expected.sha256"; then
+    fail "$tarball has no entry in SHASUMS256.txt; refusing to install."
+    rm -rf "$work_dir"; return 1
+  fi
+
+  if ! (cd "$work_dir" && verify_sha256 expected.sha256); then
+    fail "Checksum mismatch for $tarball; refusing to install."
+    rm -rf "$work_dir"; return 1
+  fi
+  ok "Checksum verified"
+
+  tar xzf "$work_dir/$tarball" -C "$work_dir"
+  extracted="$work_dir/node-${NODE_BOOTSTRAP_VERSION}-${node_os}-${node_arch}"
+  if [[ ! -x "$extracted/bin/node" ]]; then
+    fail "Extracted archive has no bin/node"
+    rm -rf "$work_dir"; return 1
+  fi
+
+  mkdir -p "$NODE_ROOT"
+  target="$NODE_ROOT/$NODE_BOOTSTRAP_VERSION"
+  rm -rf "$target.incoming"
+  mv "$extracted" "$target.incoming"
+  rm -rf "$target"
+  mv "$target.incoming" "$target"
+  ln -sfn "$NODE_BOOTSTRAP_VERSION" "$NODE_ROOT/current"
+  rm -rf "$work_dir"
+
+  node_bin="$NODE_ROOT/current/bin"
+  if ! node_is_suitable "$node_bin/node"; then
+    fail "Installed node at $node_bin/node did not run"
+    return 1
+  fi
+
+  export PATH="$node_bin:$PATH"
+  ok "Node.js $(node -v) installed with npm $(npm -v 2>/dev/null || echo '?')"
+  return 0
+}
+
+# Make the Mosaic-managed Node reachable from the next shell as well as this
+# one. Measured on a greenfield canary run: without this the install finished
+# rc=0, wrote $PREFIX/bin to ~/.profile, and the next login shell found `mosaic`
+# and then died on `env: 'node': No such file or directory` — the CLI is a Node
+# script, so a CLI on PATH without its runtime is a successful install that
+# produces a broken command.
+persist_node_on_path() {
+  persist_on_path "$NODE_ROOT/current/bin" "Mosaic-managed Node.js" || true
+}
+
+ensure_node() {
+  if command -v node &>/dev/null && node_is_suitable node; then
+    return 0
+  fi
+
+  # A previous run may have installed one that is not on this shell's PATH.
+  if node_is_suitable "$NODE_ROOT/current/bin/node"; then
+    export PATH="$NODE_ROOT/current/bin:$PATH"
+    persist_node_on_path
+    return 0
+  fi
+
+  if [[ "${MOSAIC_SKIP_NODE_BOOTSTRAP:-0}" == "1" ]]; then
+    fail "No suitable Node.js and MOSAIC_SKIP_NODE_BOOTSTRAP=1; refusing to download."
+    echo "  Install Node.js >= $NODE_MIN_MAJOR yourself, then re-run this script."
+    exit 1
+  fi
+
+  require_cmd curl
+  require_cmd tar
+
+  # sha256sum on Linux, shasum on macOS. Verification is not optional: without a
+  # checksum this would install an unauthenticated runtime.
+  if command -v sha256sum &>/dev/null; then
+    verify_sha256() { sha256sum -c --status "$1"; }
+  elif command -v shasum &>/dev/null; then
+    verify_sha256() { shasum -a 256 -c --status "$1"; }
+  else
+    fail "sha256sum or shasum required to verify the Node.js download"
+    exit 1
+  fi
+
+  if ! install_node; then
+    fail "Could not bootstrap Node.js. Install Node.js >= $NODE_MIN_MAJOR and re-run."
+    exit 1
+  fi
+  persist_node_on_path
+}
+
 # ─── preflight ────────────────────────────────────────────────────────────────
 
+ensure_node
 require_cmd node
 require_cmd npm
 
@@ -719,11 +967,7 @@ if [[ "$FLAG_CLI" == "true" ]]; then
     ensure_monorepo
     install_cli_from_source
 
-    # PATH check for npm prefix
+    ensure_prefix_on_path
-    if [[ ":$PATH:" != *":$PREFIX/bin:"* ]]; then
-      warn "$PREFIX/bin is not on your PATH"
-      dim "  Add to your shell rc:  export PATH=\"$PREFIX/bin:\$PATH\""
-    fi
   elif is_next_registry_lane; then
     info "Next mode — trying fast npm @next install from ${REGISTRY}…"
     if install_next_cli_from_registry; then
@@ -736,11 +980,7 @@ if [[ "$FLAG_CLI" == "true" ]]; then
       export MOSAIC_GATEWAY_SKIP_NPM_INSTALL=1
     fi
 
-    # PATH check for npm prefix
+    ensure_prefix_on_path
-    if [[ ":$PATH:" != *":$PREFIX/bin:"* ]]; then
-      warn "$PREFIX/bin is not on your PATH"
-      dim "  Add to your shell rc:  export PATH=\"$PREFIX/bin:\$PATH\""
-    fi
   else
     if [[ -z "$LATEST" ]]; then
       warn "Could not reach registry at $REGISTRY — skipping npm CLI."
@@ -758,11 +998,7 @@ if [[ "$FLAG_CLI" == "true" ]]; then
       ok "CLI is at or ahead of registry ($CURRENT ≥ $LATEST)."
     fi
 
-    # PATH check for npm prefix
+    ensure_prefix_on_path
-    if [[ ":$PATH:" != *":$PREFIX/bin:"* ]]; then
-      warn "$PREFIX/bin is not on your PATH"
-      dim "  Add to your shell rc:  export PATH=\"$PREFIX/bin:\$PATH\""
-    fi
   fi
 fi