@ours.network/hermes 1.2.0-nightly.1 → 1.2.0-nightly.10

package/README.md

@@ -37,7 +37,7 @@ not a background daemon that wakes a dormant agent.
 
 - Node.js ≥ 22
 - Hermes installed (`~/.hermes/` present)
-- The ours daemon: `npm i -g @ours.network/mcp@latest` (the installer does this for you)
+- The ours daemon (server): `npm i -g @ours.network/daemon` (the installer does this for you)
 
 > **Fastest path:** the one-shot [ours.network installer](../installer/README.md) sets up the
 > daemon and Hermes in one pass —
@@ -48,7 +48,7 @@ not a background daemon that wakes a dormant agent.
 
 ```sh
 npm i -g @ours.network/hermes@latest
-ours-hermes-install                 # ensures the daemon, wires the MCP server + skill only
+ours-hermes-install                 # validates the shared profile, wires the MCP server + skill
 ```
 
 That's it — the MCP server and the `ours` skill are live immediately; run **`/reload-mcp`**
@@ -56,8 +56,8 @@ in Hermes to load the `mcp_ours_*` tools. The base install asks **zero** questio
 identities or wake-on-mail — those are set up later, in-session, via the `ours` skill.
 
 `ours-hermes-install` is a thin front-door over this package's `install.sh` (below); both
-are idempotent, so re-running is always safe. Other flags: `--port`, `--hermes-dir`,
-`--skip-daemon`, `--help`.
+are idempotent, so re-running is always safe. Provision the shared profile with `ours-install client` first. Other flags:
+`--hermes-dir`, `--help`; legacy `--skip-daemon` is a no-op.
 
 ### Optional: get woken on new mail
 
@@ -82,8 +82,7 @@ re-invoke a dormant agent on background output.
 Equivalently, from a checkout you can run `bash install.sh` directly (same env knobs).
 `install.sh` is idempotent and:
 
-1. ensures `@ours.network/mcp@latest` — an existing daemon is **upgraded** (not skipped),
-   and restarted if the version changed, so a re-run is a clean upgrade;
+1. validates the installed gateway client profile before any plugin changes;
 2. removes any **legacy connector-era artifacts** an older build left behind
    (`ours-connector.env`/`.log` and a stale `ours-wake` webhook block in `config.yaml`);
 3. installs the `ours` + `writing-agent-bios` skills into
@@ -91,9 +90,9 @@ Equivalently, from a checkout you can run `bash install.sh` directly (same env k
 4. writes the `ours` MCP server into `~/.hermes/config.yaml` — **safely**: if your
    config already defines `mcp_servers:` or `platforms:`, it prints the block for
    you to merge by hand instead of risking a duplicate-key corruption;
-5. echoes the installed daemon + plugin versions so you can confirm you are on latest.
+Server installation, upgrades and lifecycle are separate administrator operations.
 
-That's the whole base install — daemon, skills, and the `ours` MCP server. Wake-on-mail
+That installs skills and the `ours` MCP client wiring. Wake-on-mail
 is enabled later, in-session, via the agent's `ours-mcp watch` tail (see *Optional: get
 woken on new mail* above); the installer writes no route, secret, or watcher.
 
@@ -104,7 +103,7 @@ Then run **`/reload-mcp`** in Hermes so it loads the `mcp_ours_*` tools.
 | var | default | purpose |
 |---|---|---|
 | `HERMES_DIR` | `~/.hermes` | config + skills root |
-| `OURS_INSTALL_SKIP_DAEMON` | — | skip the daemon step |
+| `OURS_CONFIG` | `~/.ours-client/profile.json` | complete shared gateway profile |
 
 ## Install (manual)
 

package/bin/ours-hermes-install.mjs

@@ -5,7 +5,7 @@
 //     npm i -g @ours.network/hermes
 //     ours-hermes-install
 //
-// It resolves this package's own install.sh (which ensures the ours daemon, registers the
+// It resolves this package's own install.sh (which validates the shared gateway profile, registers the
 // `ours` MCP server in ~/.hermes/config.yaml, and installs the skills) and runs it — no env-var
 // gymnastics. The MCP server + skill install immediately. Wake-on-mail is NOT set up here: the
 // agent enables it in-session by tailing `ours-mcp watch <identity>` (see the ours skill),
@@ -36,13 +36,13 @@ function help() {
 
   ours-hermes-install [options]
 
-Sets up the daemon + the ours MCP server + the skill. It asks nothing about identities or
+Requires an ours-install client profile; sets up the MCP server and skill. It asks nothing about identities or
 wake-on-mail: you enable wake in-session from your agent (bind an identity, then ask the ours
 skill to "wake me on new mail" — it tails ours-mcp watch and reacts in-session).
 
 Options:
       --hermes-dir <dir>   Hermes config+skills root (default ~/.hermes)
-      --skip-daemon        do not install/start the ours daemon
+      --skip-daemon        compatibility flag; server lifecycle is never changed
   -h, --help               show this help
 
 Idempotent: safe to re-run. After it finishes, run /reload-mcp in Hermes.`);
@@ -57,7 +57,6 @@ if (!existsSync(INSTALL)) {
 // the env vars it already understands.
 const env = { ...process.env };
 if (opts.hermesDir) env.HERMES_DIR = opts.hermesDir;
-if (opts.skipDaemon) env.OURS_INSTALL_SKIP_DAEMON = '1';
 
 const res = spawnSync('bash', [INSTALL], { stdio: 'inherit', env });
 if (res.error) { console.error(`ours-hermes-install: ${res.error.message}`); process.exit(1); }

package/install.sh

@@ -1,6 +1,6 @@
 #!/usr/bin/env bash
 # Install the ours.network plugin into Hermes:
-#   1. ensure the ours operator CLI and MCP adapter are installed; start the shared daemon
+#   1. verify the installed gateway client profile before modifying plugin configuration
 #   2. install the ours + writing-agent-bios skills into ~/.hermes/skills/
 #   3. write the `ours` MCP server into ~/.hermes/config.yaml (idempotent, never corrupts
 #      existing YAML)
@@ -10,7 +10,6 @@
 #
 # Idempotent: safe to re-run. Test/CI knobs (all optional):
 #   HERMES_DIR                 config+skills root         (default ~/.hermes)
-#   OURS_INSTALL_SKIP_DAEMON=1 skip daemon install/start
 set -euo pipefail
 
 SELFDIR="$(cd "$(dirname "$0")" && pwd)"
@@ -24,30 +23,17 @@ MANAGED_SENTINEL_END='# <<< ours.network plugin'
 
 say(){ printf 'ours-install: %s\n' "$1"; }
 
-# Ensure the ours daemon is on @latest (UPGRADE, not install-if-missing): an already-present
-# daemon must still be pulled up to the newest published version — that is the whole point of a
-# re-run. Record the CLI version before/after; start if not running, restart only if the version
-# actually changed, so the RUNNING daemon always ends on latest.
-ensure_daemon_latest(){
-  if [ "${OURS_INSTALL_SKIP_DAEMON:-}" = "1" ]; then say "skipping daemon step (OURS_INSTALL_SKIP_DAEMON=1)"; return 0; fi
-  local before after
-  before="$(ours version 2>/dev/null | grep -oE '[0-9]+\.[0-9]+\.[0-9]+' | head -1 || true)"
-  say "ensuring @ours.network/cli@latest and @ours.network/mcp@latest…"
-  npm i -g @ours.network/cli@latest @ours.network/mcp@latest
-  after="$(ours version 2>/dev/null | grep -oE '[0-9]+\.[0-9]+\.[0-9]+' | head -1 || true)"
-  if ! ours daemon status >/dev/null 2>&1; then
-    say "starting the ours daemon…"
-    if ! ours daemon start; then say "could not start the ours daemon; installation stopped."; return 1; fi
-  elif [ -n "$before" ] && [ "$before" != "$after" ]; then
-    say "operator CLI upgraded (v${before} → v${after}) — restarting its daemon…"
-    if ! ours daemon restart; then
-      say "restart failed; trying a clean start…"
-      if ! ours daemon start; then say "could not restart or start the upgraded daemon; installation stopped."; return 1; fi
-    fi
-  else
-    say "daemon already current (v${after:-unknown})."
+# Client installation never installs, starts or restarts server services.
+require_gateway_client(){
+  if ! command -v ours >/dev/null 2>&1 || ! command -v ours-mcp >/dev/null 2>&1; then
+    say "Install gateway clients with ours-install client first; ours and ours-mcp are required."
+    return 1
+  fi
+  local contract
+  if ! contract="$(ours-mcp verify-client-profile)" || [ "$contract" != "ours.gateway-client-profile-v1" ]; then
+    say "Configure the shared gateway profile with ours-install client first; plugin setup was left unchanged."
+    return 1
   fi
-  say "operator CLI: $(command -v ours) (v${after:-unknown}); MCP adapter: $(command -v ours-mcp)"
 }
 
 # Idempotent, GUARDED cleanup of legacy connector-era artifacts earlier (0.2.0/0.3.0) installers
@@ -74,8 +60,8 @@ legacy_cleanup(){
   fi
 }
 
-# --- 1) daemon (ensure @latest + restart on change) ---
-ensure_daemon_latest
+# --- 1) shared gateway client preflight ---
+require_gateway_client
 
 # --- 1b) legacy connector-era cleanup (idempotent, guarded) ---
 legacy_cleanup
@@ -95,11 +81,5 @@ HERMES_CONFIG="$HERMES_CONFIG" node "$SELFDIR/bin/hermes-config-install.mjs" ||
 }
 
 say "done. Run /reload-mcp in Hermes to load the mcp_ours_* tools."
-# --- version echo: show the user they are on latest ---
-if [ "${OURS_INSTALL_SKIP_DAEMON:-}" != "1" ]; then
-  say "versions:"
-  say "  MCP adapter: $(ours-mcp --version 2>/dev/null | head -1 || echo 'unknown')"
-  say "  plugin: $(npm ls -g @ours.network/hermes 2>/dev/null | grep -oE '@ours\.network/hermes@[0-9][0-9.]*' | head -1 || echo '@ours.network/hermes (not a global install)')"
-fi
 say "next: in your agent, bind (or create) an identity and ask the ours skill to \"wake me on new"
 say "      mail\" — it tails ours-mcp watch in-session and reacts to new mail as it arrives."

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@ours.network/hermes",
-  "version": "1.2.0-nightly.1",
+  "version": "1.2.0-nightly.10",
   "description": "Hermes (Nous Research) plugin for ours — secure agent-to-agent messaging over ADAPT. Registers the ours MCP server, bundles the ours skill, and wires in-session wake-on-mail via `ours-mcp watch` (no webhook, no external watcher).",
   "type": "module",
   "license": "FSL-1.1-Apache-2.0",

package/skills/ours/SKILL.md

@@ -108,12 +108,12 @@ existing file.
    **control-plane monitoring proxy is not available in this release** — there is no tool
    to call. See "Control plane" below.
 
-- **Configuration.** Port, state dir, broker, and GC interval are configurable
-  (env > `~/.ours/config.json` > default; port default 3050). Daemon config is
-  **host-wide and shared** — changing it restarts the daemon and drops every
-  session's binding. Never self-configure on your own initiative: surface the
-  need, explain the impact, and act only on the user's explicit yes. Details:
-  `references/configuration.md`.
+- **Configuration.** Every client selects the HTTP gateway through
+  `~/.ours-client/profile.json` (or the whole-profile `OURS_CONFIG` override).
+  `serverUrl` is the base URL; daemon and Cowork use `/daemon` and `/cowork`
+  beneath it. Import the issued credential with `ours-install client`.
+  Local daemon ports, state directories, and Unix sockets are not client routes.
+  Missing or invalid configuration fails closed. See `references/configuration.md`.
 
 ## ours-fleet — persistent harness roles
 
@@ -412,9 +412,8 @@ release, and do not improvise a substitute. Per-identity wake-on-mail is a **dif
 feature and still works; it is described above.
 ## Notes
 
-- Identities and their state (contacts, history, keys) persist under the daemon's state dir
-  (`OURS_STATE_DIR`, default `~/.ours`) and survive restarts. The daemon is a singleton
-  shared by all your Hermes agents and sessions on this host.
+- Identity state is stored on the selected server and survives client restarts.
+  Clients access it through the gateway; no server state directory is required locally.
 - Inbound messages from unknown (non-contact) senders are rejected — only peers added via an
   invite handshake, same-host agents under the same Human identity, or registrar-verified
   local-contact-book introductions can reach you.

package/skills/ours/references/configuration.md

@@ -1,26 +1,24 @@
-# ours daemon configuration
+# Shared gateway client configuration
 
-ours-mcp is a client of one already-running shared daemon. The operator CLI owns
-configuration and lifecycle:
+Fleet, ours CLI, MCP and harness hooks select one server through
+`~/.ours-client/profile.json`. `serverUrl` names the HTTP(S) gateway, with
+optional base path. Daemon operations use `/daemon`; Cowork uses `/cowork`.
+The profile also pins `expectedInstanceId` and an absolute `credentialPath`.
+An optional `endpoint` must equal `serverUrl + "/daemon"`.
 
-```sh
-ours config show --json
-ours config setup --port 3050 --state-dir "$HOME/.ours"
-ours daemon start
-ours daemon status --json
-```
+Import a prepared private profile with `ours-install client --config /absolute/profile.json`.
+Use `ours config show --json` to inspect connection metadata without reading the token.
+Keep the profile directory mode 0700 and files mode 0600. `OURS_CONFIG` may
+select the same whole profile for all clients. Remove legacy `OURS_PORT`,
+`OURS_STATE_DIR`, `OURS_API_TOKEN` and `OURS_DAEMON_ID` overrides.
 
-The MCP adapter uses the published SDK's coherent selection. The wholly default
-selection is port 3050 with state directory `~/.ours`. For another daemon, set
-`OURS_CONFIG`, or set matching `OURS_PORT` and `OURS_STATE_DIR`. A token or
-endpoint selection must be paired with its state directory. The daemon's
-`/state-dir` response is verified before credentials are sent.
+Missing profiles, rejected credentials and unavailable gateways fail closed.
+Clients never read daemon state, start a daemon, or try a local port/socket.
+Hooks use the same gateway and return a harmless no-op on failure. Their
+application-local identity/session files do not select another server.
 
-The adapter never starts a daemon and never falls back to an embedded one.
-`OURS_INSTANCE`, `--application`, and old ours-mcp daemon variables are errors.
-Do not add a duplicate MCP registration; the managed command is simply
-`ours-mcp proxy`.
-
-Changing daemon configuration or restarting the shared daemon affects every
-connected application. Explain that blast radius and obtain the user's consent
-before making operator-level changes.
+For full-stack setup, a custom public port (such as 4050), systemd propagation
+and migration, see the [gateway setup guide](https://github.com/adapt-toolkit/ours-network/blob/prerelease/packages/installer/GATEWAY_SETUP.md).
+The coordinated gateway changes require approved matching package releases;
+current npm nightly is not automatically evidence that these changes are installed.
+Server administration uses `ours-install server` on the server host.