multica_agent_host: op shim reads the service account token (WDWL-37)

Deliver vault_op_service_account_claude to the agent hosts and install a
role-owned ~multica/.local/bin/op shim that exports OP_SERVICE_ACCOUNT_TOKEN
to /usr/bin/op only. op run / op plugin run get env -u spliced after "--"
so the child never sees the token. The daemon unit is untouched (WDWL-18).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

ansible/roles/multica_agent_host/README.md

@@ -38,7 +38,7 @@ otherwise clean at the `production` profile.
 | `multica` CLI, pinned, from the GitHub release with `checksums.txt` verification | yes |
 | Agent CLIs (`claude`, `hermes`, …) | yes |
 | 1Password CLI (`op`), from 1Password's signed apt/yum repository | yes (`multica_op_install`) |
-| 1Password service account token, as a 0600 file for the daemon user | opt-in (`multica_op_deliver_token`); not exported to the daemon |
+| 1Password service account token, as a 0600 file for the daemon user, plus an `op` shim that reads it | opt-in (`multica_op_deliver_token`); not exported to the daemon |
 | `workspaces_root`, on the intended mount, writable by the daemon user | yes |
 | Multica authentication (`multica login --token`) | yes |
 | CLI config (`server_url`, `app_url`, `workspaces_root`, `device_name`, …) | yes |
@@ -241,6 +241,7 @@ See `defaults/main.yml` for the annotated list. The ones you will actually set:
 | `multica_op_install` | `true` | 1Password CLI from 1Password's apt/yum repo. |
 | `multica_op_version` | `""` | `""` tracks the repo (upgrades on every run); set e.g. `2.41.1` to pin. |
 | `multica_op_deliver_token` | `false` | Write `multica_op_service_account_token` to `multica_op_token_file`. |
+| `multica_op_shim` | `true` | With the token delivered, install the `op` shim at `multica_op_shim_path`. |
 
 ### Adding an agent CLI
 
@@ -399,13 +400,13 @@ The role can deliver the token, but by default does not:
 
 ```yaml
 multica_op_deliver_token: true
-multica_op_service_account_token: "{{ vault_op_service_account_tokens[inventory_hostname_short] }}"
+multica_op_service_account_token: "{{ vault_op_service_account_claude }}"
 ```
 
 writes it to `multica_op_token_file`
 (`~multica/.config/op-service-account/token`, directory 0700, file 0600,
-owned by the daemon user). That is all it does. **The role does not put the
-token into the daemon's environment**, for two reasons:
+owned by the daemon user) and installs an `op` shim. **The role does not put
+the token into the daemon's environment**, for two reasons:
 
 1. The daemon's systemd user unit is written and owned by the multica CLI
    (`multica daemon autostart`). The role must not template or edit it, or it
@@ -413,16 +414,35 @@ token into the daemon's environment**, for two reasons:
 2. Anything in the daemon's environment is inherited by every agent process
    it spawns, where a casual `env` or `printenv` puts it in a transcript.
 
-How the token reaches `op` is a separate decision, not yet made. Until then
-an agent can use it explicitly:
+`op` reads the token only from `OP_SERVICE_ACCOUNT_TOKEN`; it has no
+token-file option. So the role installs a shim at `multica_op_shim_path`
+(`~multica/.local/bin/op`, root-owned 0755), which the CLI-written unit
+already puts first on `PATH`. The shim:
+
+- reads the token file and exports `OP_SERVICE_ACCOUNT_TOKEN` to
+  `multica_op_binary` (`/usr/bin/op`) only, then `exec`s it. Agents call plain
+  `op run` / `op inject` / `op read` and the token never enters their shell;
+- fails closed (exit 78, no token in the message) when the file is missing
+  or empty;
+- for `op run` and `op plugin run`, splices `/usr/bin/env -u
+  OP_SERVICE_ACCOUNT_TOKEN` in after the first `--`. `op run` otherwise
+  hands its whole environment, token included, to the command it starts.
+  Because it needs that `--` to find the command, the shim refuses
+  `op run <command>` without one (exit 64).
+
+The play checks, as the daemon user, that `op` resolves to the shim and that
+the token is not in that environment.
+
+The shim keeps the token out of transcripts by accident; it is not a
+boundary. Any process running as the daemon user can read the token file on
+purpose. What actually limits exposure is the service account's vault access
+in 1Password: keep it read-only. Check it as the daemon user:
 
 ```bash
-OP_SERVICE_ACCOUNT_TOKEN="$(cat ~/.config/op-service-account/token)" op run -- <cmd>
+op vault list --permission read_items     # vaults it can read
+op vault list --permission edit_items     # should be empty
 ```
 
-`op` reads the token only from `OP_SERVICE_ACCOUNT_TOKEN`; it has no
-token-file option.
-
 ## Known gaps
 
 - **Provider CLI login.** Unavoidable for subscription auth.
@@ -444,7 +464,7 @@ token-file option.
 | `multica_auth` | `multica login` |
 | `multica_daemon` | autostart + start |
 | `multica_verify` | the assertions |
-| `multica_op` | 1Password CLI, and the token file when enabled |
+| `multica_op` | 1Password CLI, and the token file and shim when enabled |
 | `multica` | all of the above |
 
 Preflight and the service-account step are tagged `always`: they build

ansible/roles/multica_agent_host/defaults/main.yml

@@ -158,13 +158,18 @@ multica_op_apt_base_url: "https://downloads.1password.com/linux/debian"
 multica_op_rpm_base_url: "https://downloads.1password.com/linux/rpm/stable"
 
 # Service account token delivery. Off by default. When on, the token is written
-# to multica_op_token_file (0600, owned by the daemon user) and nowhere else.
-# The role does NOT export it into the daemon environment; see the README.
-# Supply the VALUE from the playbook, like multica_pat, e.g.
-#   multica_op_service_account_token: "{{ vault_op_service_account_tokens[inventory_hostname_short] }}"
+# to multica_op_token_file (0600, owned by the daemon user) and read from there
+# by the op shim below. The role does NOT export it into the daemon
+# environment; see the README. Supply the VALUE from the playbook, e.g.
+#   multica_op_service_account_token: "{{ vault_op_service_account_claude }}"
 multica_op_deliver_token: false
 multica_op_service_account_token: ""
 multica_op_token_file: "{{ multica_user_home }}/.config/op-service-account/token"
+# With the token delivered, install a shim at multica_op_shim_path (first on
+# the daemon unit's PATH) that exports the token to multica_op_binary only.
+multica_op_shim: true
+multica_op_shim_path: "{{ multica_user_home }}/.local/bin/op"
+multica_op_binary: /usr/bin/op
 
 # -----------------------------------------------------------------------------
 # Node.js (installed only when a selected agent CLI needs npm)

ansible/roles/multica_agent_host/tasks/op_token.yml

@@ -35,3 +35,47 @@
     mode: "0600"
   no_log: true
   become: true
+
+# --- op shim -----------------------------------------------------------------
+# `op` reads the token only from OP_SERVICE_ACCOUNT_TOKEN. The shim, first on
+# the daemon unit's PATH, exports it for the real binary alone, so agents call
+# plain `op` and the token never enters their shell environment. The daemon
+# unit itself stays the CLI's (WDWL-18).
+
+- name: 1Password token | Create the daemon user's bin directory
+  ansible.builtin.file:
+    path: "{{ multica_op_shim_path | dirname }}"
+    state: directory
+    owner: "{{ multica_user }}"
+    group: "{{ multica_group }}"
+    mode: "0755"
+  become: true
+  when: multica_op_shim | bool
+
+- name: 1Password token | Install the op shim
+  ansible.builtin.template:
+    src: op-shim.sh.j2
+    dest: "{{ multica_op_shim_path }}"
+    owner: root
+    group: root
+    mode: "0755"
+  become: true
+  when: multica_op_shim | bool
+
+# As the daemon user, through its own environment: plain `op` must resolve to
+# the shim, and the token must not be in that environment.
+- name: 1Password token | Check op resolves to the shim as the daemon user
+  ansible.builtin.shell:
+    cmd: 'command -v op; env | grep -c "^OP_SERVICE_ACCOUNT_TOKEN=" || true'
+  become: true
+  become_user: "{{ multica_user }}"
+  environment: "{{ _multica_user_env }}"
+  register: _multica_op_shim_check
+  changed_when: false
+  check_mode: false
+  failed_when: >-
+    _multica_op_shim_check.rc != 0
+    or _multica_op_shim_check.stdout_lines != [multica_op_shim_path, '0']
+  when:
+    - multica_op_shim | bool
+    - not ansible_check_mode

ansible/roles/multica_agent_host/templates/op-shim.sh.j2

@@ -0,0 +1,55 @@
+#!/bin/sh
+# {{ ansible_managed }}
+#
+# 1Password CLI shim for the Multica daemon user (multica_agent_host role,
+# WDWL-37). Reads the service account token from a 0600 file and exports it
+# to {{ multica_op_binary }} only, so it never sits in an agent's shell
+# environment.
+token_file="{{ multica_op_token_file }}"
+
+if [ ! -r "$token_file" ]; then
+    echo "op: no 1Password service account token at $token_file" >&2
+    exit 78
+fi
+IFS= read -r OP_SERVICE_ACCOUNT_TOKEN < "$token_file"
+if [ -z "$OP_SERVICE_ACCOUNT_TOKEN" ]; then
+    echo "op: 1Password service account token file $token_file is empty" >&2
+    exit 78
+fi
+export OP_SERVICE_ACCOUNT_TOKEN
+
+# `op run` (and `op plugin run`) hand their own environment, token included,
+# to the command they start. Find the subcommand, skipping global flags that
+# take a value, and stop at the first `--`.
+sub1=""; sub2=""; skip=0; help=0
+for a in "$@"; do
+    if [ "$skip" = 1 ]; then skip=0; continue; fi
+    case "$a" in
+        --) break ;;
+        -h|--help) help=1 ;;
+        --account|--config|--encoding|--format|--session) skip=1 ;;
+        -*) ;;
+        *) if [ -z "$sub1" ]; then sub1=$a; elif [ -z "$sub2" ]; then sub2=$a; fi ;;
+    esac
+done
+
+if [ "$sub1" = run ] || { [ "$sub1" = plugin ] && [ "$sub2" = run ]; }; then
+    # Rebuild the arguments with `env -u` spliced in after the first `--`, so
+    # the child starts without the token.
+    n=$#; i=0; spliced=0
+    while [ "$i" -lt "$n" ]; do
+        a=$1; shift
+        set -- "$@" "$a"
+        if [ "$spliced" = 0 ] && [ "$a" = "--" ]; then
+            set -- "$@" /usr/bin/env -u OP_SERVICE_ACCOUNT_TOKEN
+            spliced=1
+        fi
+        i=$((i + 1))
+    done
+    if [ "$spliced" = 0 ] && [ "$help" = 0 ]; then
+        echo "op: use 'op run [flags] -- <command>'; the shim needs the '--' to keep the service account token out of <command>" >&2
+        exit 64
+    fi
+fi
+
+exec "{{ multica_op_binary }}" "$@"