@ory/goose 0.14.0 → 1.0.0

package/README.md

@@ -2,9 +2,12 @@
 
 Security and developer experience for [Goose](https://github.com/block/goose) (Block's open-source coding agent), powered by [Ory](https://ory.com).
 
-**Security.** Goose runs real actions on your machine — editing files, running shell commands, calling APIs. The plugin gives every session a verifiable identity (you sign in once; Goose and any sub-agents it spawns each get their own), checks every tool call against permissions you control, and records each decision as an audit trace you can ship to your observability stack. It starts in watch mode so nothing is blocked on day one, and if Ory is ever unreachable it steps aside rather than locking you out.
+Goose runs real actions on your machine — editing files, running shell commands, calling APIs. This plugin gives those actions an identity, a permission check, and an audit trail.
 
-**Developer experience.** A single command installs the plugin and walks you through connecting — choose Ory Network, a local Docker stack, or audit-only, and it wires up the project, sign-in client, login, and permissions for you. It also helps you build Ory into your own app: ask in plain language to scaffold login, registration, and recovery pages, run a local Ory, or manage identities and permissions through the bundled MCP server.
+One command installs two independent halves:
+
+- **Developer experience** — the Ory skill catalog, the local Ory dev stack, and activity logging. No account, no keys, nothing to configure: it works the moment it's installed.
+- **Ory Agent Security** — browser sign-in, brokered permission checks before tool calls, and the delegation trail. Opt-in connection details from the [Ory Console](https://console.ory.sh) switch it on, and it takes nothing away from the half above. See [Connect to Ory Agent Security](#connect-to-ory-agent-security).
 
 Verified against the `goose` binary (v1.39.0): the plugin layout, the `hooks.json` schema, and the discovery paths all match Goose's shell-hook plugin contract (see [block/goose#9304](https://github.com/block/goose/pull/9304)).
 
@@ -17,117 +20,121 @@ Verified against the `goose` binary (v1.39.0): the plugin layout, the `hooks.jso
 
 ## Get started
 
-Run one command. It installs the plugin and walks you through connecting:
+Run one command:
 
 ```bash
 npx -y -p @ory/goose ory-goose install
 ```
 
-The installer drops a self-contained plugin at `<project>/.agents/plugins/ory/`, where Goose finds it on its own at startup — no `goose configure` step and nothing else in your Goose config gets touched. Then it opens a guided setup **in your browser** where you pick how to connect — the default takes just a click:
-
-- **Ory Network** *(default)* — sign in, or create a free account, in your browser. The project, keys, permissions, and login are all set up for you. Nothing to configure by hand.
-- **Local** — run a complete Ory on your laptop with Docker. No account, no signup, no keys. Great for trying it out.
-- **Audit-only** — skip Ory entirely and just log what Goose does.
+The installer drops a self-contained plugin at `<project>/.agents/plugins/ory/`, where Goose finds it on its own at startup — no `goose configure` step and nothing else in your Goose config gets touched. Add `--global` to install for every project (at `~/.agents/plugins/ory/`).
 
-> No browser available (CI, SSH, headless)? The same walkthrough runs right in your terminal instead — or force it with `--no-web`.
-
-That's it. Confirm everything landed with:
+Confirm everything landed with:
 
 ```bash
 npx -y -p @ory/goose ory-goose status
 ```
 
-`status` is your one-stop check: what's configured, who's signed in, which tools are covered by permissions, whether the plugin is registered (project and global), and recent activity.
+`status` is your one-stop check: what's configured, who's signed in, which tools are covered by permissions, whether the plugin is registered (project and global), and recent activity. Until you connect Agent Security, the identity and permission rows say so and name what's missing.
+
+## Skills and commands
+
+Installing the plugin drops the full Ory playbook catalog into Goose. **Skills** are vetted playbooks Goose runs when you ask for them in chat — plain language is enough.
+
+| Skill | What it does for you |
+|---|---|
+| `ory-auth-setup` | Adds a complete auth system to your app — login, registration, recovery, verification, settings — on [Ory Elements](https://github.com/ory/elements) |
+| `ory-login-flow` | Builds just the pages, wired to Ory's self-service flows |
+| `ory-social-login` | "Sign in with…" for Google, GitHub, Apple, Microsoft, Discord, Slack, GitLab, Facebook |
+| `ory-local-dev` | Develops and tests login/permission flows against a local Ory — no project, no account, offline |
+| `ory-permissions-onboarding` | Walks a fresh install from observe mode to enforced per-tool permissions without getting blocked |
+| `ory-build-agent` | Drops `@ory/argus` into an agent *you* own — Claude Agent SDK, OpenAI Agents, Mastra, Vercel AI, LangGraph/PydanticAI |
+| `ory-build-integration` | Wires Ory into your app: Action webhooks, JWT validation at a gateway, live event streams |
+| `ory-contribute-integration` | Authors and submits an integration to the public `ory/integrates` registry |
+| `ory-e2b-sandbox` | Scaffolds an E2B sandbox template that boots with this plugin preinstalled |
+| `ory-temporal-worker` | Scaffolds a Temporal TypeScript worker where every Activity is authenticated, authorized, and audited |
+
+The local stack has its own playbooks — ask for them by name:
+
+| Command | What it does |
+|---|---|
+| `ory-local-up` | Starts a local Ory (Identities, OAuth2, Permissions) in Docker and seeds a test user — it prints the email + password to sign in with |
+| `ory-local-down` | Stops it, keeping your data volumes |
+| `ory-temporal-up` | Starts a local Temporal dev server for the `ory-temporal-worker` scaffold |
 
-Re-run install with `--reconfigure` to change your connection later, `--no-configure` to skip the wizard, or `--global` to install for every project (at `~/.agents/plugins/ory/`).
+The local stack runs entirely on your laptop: Ory APIs at `http://localhost:4000`, a login UI on `:4455` (not `:3000`, to dodge Next.js port clashes), and the Ory Console on `:4100`.
+
+So: ask Goose *"add Ory login to this app"* and it scaffolds the pages, starts a local Ory, and wires them together.
 
 ## What you get
 
-Once connected, every tool Goose runs is governed by Ory — three things happen automatically:
+Out of the box, every tool Goose runs produces a privacy-safe structured activity event in the unified local log.
+
+Once you connect to Ory Agent Security, two more things happen automatically:
 
-- **Who's driving.** You sign in once in your browser (a standard, secure browser login); Goose gets its own separate identity and registers it automatically on first run. No tokens to copy around, and the "who did what" trail stays queryable later.
-- **What it's allowed to do.** Before a tool runs, Ory checks whether it's permitted. It starts in **watch mode** — nothing is blocked, you just *see* what would be — so it never gets in your way on day one.
-- **A record of everything.** Every decision (allowed, denied, skipped) is logged as a trace you can send to Jaeger, Honeycomb, Grafana, or just a file.
+- **Who's driving.** You sign in once in your browser; Goose (and any sub-agents it spawns) each get their own identity. No tokens to copy around, and the "who acted on whose behalf" trail stays queryable later.
+- **What it's allowed to do.** Before a tool runs, Ory checks whether it's permitted. It starts in **observe mode** — nothing is blocked, you just *see* what would be — so it never gets in your way on day one.
 
-Under the hood, Goose runs the plugin as a small program for each thing it does, handing it the event and reading back a decision. The `hooks.json` file maps that: **SessionStart** signs you in, **PreToolUse** (the one point that can block) runs the permission check, and the rest — **PostToolUse, BeforeReadFile, AfterFileEdit, BeforeShellExecution, Stop, SessionEnd** — just record activity. If Ory is ever unreachable, the plugin gets out of the way and lets Goose keep working, so it can't lock you out.
+Under the hood, Goose runs the plugin as a small program for each thing it does, handing it the event and reading back a decision. The `hooks.json` file maps that: **SessionStart** signs you in (once connected), **PreToolUse** (the one point that can block) runs the permission check, and the rest — **PostToolUse, BeforeReadFile, AfterFileEdit, BeforeShellExecution, Stop, SessionEnd** — just record activity. If Ory is ever unreachable, the plugin gets out of the way and lets Goose keep working, so it can't lock you out.
 
 ### See what's happening
 
-Everything the plugin does is observable out of the box — no configuration required:
+Everything the plugin does is observable out of the box:
 
-- **Status at a glance.** `npx -y -p @ory/goose ory-goose status` shows what's configured, who's signed in, how many built-in tools your permissions cover, and the most recent tool-call activity.
-- **Live dashboard.** `npx -y -p @ory/goose ory-goose dashboard` opens the same picture in your browser — configuration, identities, permission coverage, Ory service health, and the latest tool-call activity, all refreshing live. From there you can flip **enforcement** on or off, toggle **user login**, and use **Change stack** to reconnect to a different Ory — no CLI required.
-- **Live traces.** Every tool call is recorded as an OpenTelemetry-style span. Watch them stream as the agent works:
+- **Activity log.** Privacy-safe activity is always appended to `~/.config/ory-agent-plugins/goose/ory-agent-debug.log`. View events, decisions, and errors live with:
 
   ```bash
   npx -y -p @ory/goose ory-goose watch
   ```
 
-  Spans are also written to `~/.config/ory-agent-plugins/goose/ory-agent-trace.ndjson` (NDJSON, one span per line) — tail that file, or point `OTEL_EXPORTER_OTLP_ENDPOINT` at a collector to ship them straight to Jaeger, Honeycomb, or Grafana.
-- **Debug log.** For a verbose play-by-play, set `ORY_AGENT_DEBUG=true`; structured logs land in `~/.config/ory-agent-plugins/goose/ory-agent-debug.log`.
+  Set `ORY_AGENT_LOG_FILE` to override the path; set it empty to disable file persistence.
+- **Live debug.** Launch Goose with `ORY_AGENT_DEBUG=true` to add verbose local diagnostics, including raw shell commands, to the watched log and stderr. Secrets are recursively redacted; pass `watch --json` for NDJSON.
 
 ### Ready to enforce?
 
-When the watch-mode logs look right, turn on blocking with one command (setup already granted you the built-in tools):
+Once connected, the deny posture lives on the Ory project: when the observe-mode activity looks right, an admin promotes it to **enforce** in the **Ory Console** (Agent Security). Every session reads that posture live, so nothing has to be reinstalled.
 
 ```bash
-npx -y -p @ory/goose ory-goose permissions enforce
+npx -y -p @ory/goose ory-goose permissions   # what the project grants, and the live mode
 ```
 
-Now a denied tool is actually blocked and Goose shows why. Go back to watch mode anytime with `permissions observe`. Prefer clicking? The dashboard has the same **Enforce** switch — flip it on, or back to watch mode, without touching the CLI. Use `permissions status` to see what's covered and `permissions bootstrap` to (re-)grant the built-in tools.
-
-## Also: add login to your own app
-
-Beyond securing Goose, the plugin helps you build Ory into whatever you're working on — backed by a local Ory, so it needs nothing but Docker. Ask Goose in plain language, e.g. *"add Ory auth to this app"*, and it scaffolds the login, registration, recovery, and settings pages (using [Ory Elements](https://github.com/ory/elements)) wired to a local Ory — no signup or keys needed.
+Then a denied tool is actually blocked and Goose shows why.
 
-These are **skills** — vetted playbooks Goose runs when you ask for them in chat. Start a local Ory with the `ory-local-up` skill (it prints a test email + password to sign in with) and tear it down with `ory-local-down`. More skills cover `ory-auth-setup`, `ory-login-flow`, `ory-social-login` (Google, GitHub, Apple…), `ory-permissions-onboarding`, and more.
+## Connect to Ory Agent Security
 
-## Configure by hand (CI / advanced)
+Copy the connection details from the [Ory Console](https://console.ory.sh) under **Agent Security**:
 
-The guided setup covers most people. For scripted or CI setups, or to point at an existing Ory Network project, configure directly. Settings are saved to `~/.config/ory-agent-plugins/config.json` and shared across all your Ory agent plugins; environment variables win when both are set.
+| Value | Flag | Environment variable |
+|---|---|---|
+| Project URL | `--project-url` | `ORY_PROJECT_URL` |
+| Agent Security URL | `--agent-security-url` | `ORY_AGENT_SECURITY_URL` |
+| Sign-in client id override (default `ory-agent-security-login`) | `--oauth2-client-id` | `ORY_OAUTH2_CLIENT_ID` |
 
 ```bash
 npx -y -p @ory/goose ory-goose configure \
   --project-url https://<slug>.projects.oryapis.com \
-  --oauth2-client-id <sign-in client id>
+  --agent-security-url https://agents.console.ory.com
 ```
 
-Goose's own identity registers itself automatically on first run — nothing to create. The `--oauth2-client-id` is the one piece browser sign-in needs (the guided setup makes it for you). The interactive user login runs every session (always on, non-blocking) and only needs `ORY_OAUTH2_CLIENT_ID` to complete the browser flow. For logging-only with no checks, use `--audit-only`. The matching env vars are `ORY_PROJECT_URL`, `ORY_OAUTH2_CLIENT_ID`, and `ORY_AGENT_API_KEY`.
+`install` accepts the same flags, so you can register the plugin **and** connect in one shot (`install --project-url <URL> --agent-security-url <URL>`). The login client defaults to `ory-agent-security-login`; custom deployments can override it with `--oauth2-client-id`. Existing configurations fall back to the project URL when the Agent Security URL is unset. To turn sign-in and checks back off later, use `configure --disconnect`.
 
-<details>
-<summary>Create the sign-in client by hand</summary>
-
-The guided setup normally does this. To do it yourself, create a **public** OAuth2 client (no secret) listing all four loopback URLs — the plugin tries each in turn so sign-in survives a busy port:
-
-```bash
-ory create oauth2-client --project <project-id> \
-  --name "Ory Agent Security · user login (PKCE)" \
-  --grant-type authorization_code,refresh_token \
-  --response-type code \
-  --scope openid,offline_access \
-  --token-endpoint-auth-method none \
-  --redirect-uri http://127.0.0.1:47823/callback \
-  --redirect-uri http://127.0.0.1:47824/callback \
-  --redirect-uri http://127.0.0.1:47825/callback \
-  --redirect-uri http://127.0.0.1:47826/callback
-```
+**Ory Network or OEL.** Either works. For an Ory Network project the URL is `https://<slug>.projects.oryapis.com`; for a self-hosted **Ory Enterprise License** deployment, point `--project-url` at that deployment's base URL and use the sign-in client id from its Agent Security configuration. Everything downstream — sign-in, checks, delegation — is identical.
 
-Pass the resulting id to `configure --oauth2-client-id`. Running headless with a session token already? Set `ORY_USER_SESSION_TOKEN` and skip the browser step entirely.
+**There is nothing else for you to create.** The sign-in client, the permission model, the per-tool grants and blocks, and the observe/enforce posture are all provisioned in the Console by someone with project access. At runtime the plugin only *reads* permissions — it has no write path into your project, which is why installing it needs no workspace privilege. Each Goose session registers its own identity automatically on first use.
 
-</details>
+Sign-in runs at the start of every session and never blocks — a declined, skipped, or timed-out login simply leaves that session without a user identity.
 
-With nothing configured, the plugin still loads and runs in **pass-through mode**: skills and logging work, but no checks run and nothing is blocked. Perfectly fine if you only want the app-building features.
+Settings are saved to `~/.config/ory-agent-plugins/config.json` and shared across all your Ory agent plugins; environment variables win when both are set. Running headless with an OAuth2 access token already? Set `ORY_USER_OAUTH2_TOKEN` and the browser step is skipped entirely.
 
 ## Commands
 
 ```
-ory-goose install | uninstall        Install/remove; --reconfigure re-runs setup, --no-web forces the terminal wizard, --no-configure skips it, --global installs everywhere
+ory-goose install | uninstall        Install (add --project-url to also connect Agent Security) / remove; --global installs for every project
 ory-goose status                     Show configuration, identities, permission coverage, registration, recent activity
-ory-goose dashboard                  Open the live browser dashboard (status + service health + Change stack)
-ory-goose watch                      Tail the live trace stream (OTel spans)
-ory-goose permissions <cmd>          status | bootstrap | observe (watch) | enforce (block)
-ory-goose configure <flags>          Point at a project by hand (--project-url, --oauth2-client-id, --audit-only)
+ory-goose permissions <cmd>          status  (read-only; grants + posture live in the Ory Console)
+ory-goose configure <flags>          Connect a project (--project-url) or --disconnect
 ory-goose agent <status|unregister>  Manage Goose's own auto-created identity
+ory-goose local <up|down|status|…>   Run / manage a local Ory in Docker
+ory-goose version                    Print plugin, core, and Node versions (--json for machine-readable)
 ```
 
 All prefixed with `npx -y -p @ory/goose`.
@@ -135,16 +142,14 @@ All prefixed with `npx -y -p @ory/goose`.
 ## Troubleshooting
 
 - **The local Ory fails to start** — make sure Docker is running and ports `4000`, `4100`, `4455`, and `16686` are free.
-- **Browser sign-in loops** — reset with `ory-goose agent unregister` and try again.
-- **Install ran but never asked how to connect** — you're almost certainly on a stale `npx` cache. `npx -p @ory/goose` (no version pin) reuses a previously-downloaded copy instead of re-resolving to the latest, so an older CLI whose install predates the setup wizard can run while you believe you're on the current release. The install banner prints the running version; confirm it with `npx -y -p @ory/goose ory-goose version`. To force the current release, clear the cache and reinstall: `rm -rf ~/.npm/_npx` then `npx -y -p @ory/goose ory-goose install`. Pinning an exact version (`@ory/goose@<version>`) also bypasses the cached copy.
+- **Browser sign-in loops** (after connecting) — reset with `ory-goose agent unregister` and try again.
+- **Running an older CLI than expected** — `npx -p @ory/goose` (no version pin) reuses a previously-downloaded copy instead of re-resolving to the latest. The install banner prints the running version; confirm it with `npx -y -p @ory/goose ory-goose version`. To force the current release, clear the cache and reinstall: `rm -rf ~/.npm/_npx` then `npx -y -p @ory/goose ory-goose install`. Pinning an exact version (`@ory/goose@<version>`) also bypasses the cached copy.
 - **Goose isn't picking up the plugin** — confirm the tree at `<project>/.agents/plugins/ory/` exists (or re-run the installer), and restart Goose.
-- **Want to see what's happening** — `npx -y -p @ory/goose ory-goose status` for a snapshot, `npx -y -p @ory/goose ory-goose watch` for the live trace stream, or set `ORY_AGENT_DEBUG=true` for a verbose log. Traces and logs live under `~/.config/ory-agent-plugins/goose/` (see [See what's happening](#see-whats-happening)).
 
 ## Learn more
 
 - [Ory documentation](https://www.ory.com/docs/) · [Ory Console](https://console.ory.sh) · [Ory Elements](https://github.com/ory/elements)
 - [Goose repository](https://github.com/block/goose)
-- [Repo README](../../README.md) and [AGENTS.md](../../AGENTS.md) — full env-var and permission-mode reference
 
 ## License
 

package/dist/cli/main.js

@@ -57,6 +57,11 @@ const PLUGIN_NAME = "ory";
 const PACKAGE_ROOT = path.resolve(__dirname, "..", "..");
 function main() {
     const [command, ...args] = process.argv.slice(2);
+    // `--help` after a matched subcommand: print usage, do nothing else.
+    // Without this the switch below ignores the flag and `install --help`
+    // performs a real install (#221).
+    if ((0, argus_1.shouldPrintHelp)(command, args))
+        return help();
     switch (command) {
         case "install":
             (0, argus_1.beginDeferNextSteps)();
@@ -68,7 +73,10 @@ function main() {
             break;
         case "uninstall":
             uninstall(args);
-            (0, argus_1.clearCredentialsForUninstall)().then(() => process.exit(0), (err) => {
+            (0, argus_1.clearCredentialsForUninstall)({
+                harness: "goose",
+                purge: args.includes("--purge"),
+            }).then(() => process.exit(0), (err) => {
                 console.error(err.message ?? err);
                 process.exit(1);
             });
@@ -77,7 +85,7 @@ function main() {
             (0, argus_1.runConfigureCommand)("ory-goose", args);
             break;
         case "agent":
-            (0, argus_1.runAgentCommand)("ory-goose", args).then((code) => process.exit(code), (err) => {
+            (0, argus_1.runAgentCommand)("ory-goose", "goose", args).then((code) => process.exit(code), (err) => {
                 console.error(err.message ?? err);
                 process.exit(1);
             });
@@ -97,17 +105,11 @@ function main() {
                 process.exit(1);
             });
             break;
-        case "local":
-            (0, argus_1.runLocalCommand)("ory-goose", args).catch((err) => {
-                console.error(err.message ?? err);
-                process.exit(1);
-            });
-            break;
         case "watch":
-            (0, argus_1.runWatchCommand)("goose", args);
+            (0, argus_1.runWatchCommand)("ory-goose", "goose", args).then((code) => process.exit(code), (err) => { console.error(err.message ?? err); process.exit(1); });
             break;
-        case "dashboard":
-            (0, argus_1.runDashboardCommand)("ory-goose", "goose", args).then((code) => process.exit(code), (err) => {
+        case "local":
+            (0, argus_1.runLocalCommand)("ory-goose", args).catch((err) => {
                 console.error(err.message ?? err);
                 process.exit(1);
             });
@@ -173,11 +175,11 @@ Commands:
   uninstall [--global]   Remove the Ory plugin
   configure              Set or view Ory project URL and API key
   agent <cmd>            Manage the agent's OAuth2 (DCR) identity (status, unregister)
-  permissions <cmd>      Manage permission mode and tool permissions (status, bootstrap, observe, enforce)
+  permissions            Show the live permission mode and per-tool coverage
   setup [--project-dir]  Write the plugin tree directly (advanced)
   status                 Show plugin status and configuration
-  dashboard            Open the Ory Agent dashboard (system health + configuration)
-  watch [trace-file]     Tail the trace stream (OTel spans) live
+  watch [--json] [--lines <count>]
+                         Follow the live activity/debug log
   version                Show version and the ory-agent-plugins build commit
   local <cmd>            Manage local Ory dev environment (up, down, status, seed, ...)
 

package/dist/cli/setup.d.ts

@@ -7,10 +7,10 @@
  *   - `~/.agents/plugins/<name>/`         (user-global, with --global)
  *
  * Each plugin dir holds a `plugin.json` manifest and `hooks/hooks.json`
- * describing the hook commands. We render both fresh at install time so the
- * version pin baked into the hook command always matches the installed
- * package. No `goose configure`, config edit, or MCP registration is needed
- * for the gate — discovery is purely directory-based.
+ * describing the hook commands. We render both fresh at install time because
+ * the hook command points at the runtime shim resolved by this install. No
+ * `goose configure`, config edit, or MCP registration is needed for the gate —
+ * discovery is purely directory-based.
  *
  * The plugin dir layout (plugin.json + hooks/hooks.json) and discovery paths
  * are verified against goose v1.39.0.

package/dist/cli/setup.js

@@ -8,10 +8,10 @@
  *   - `~/.agents/plugins/<name>/`         (user-global, with --global)
  *
  * Each plugin dir holds a `plugin.json` manifest and `hooks/hooks.json`
- * describing the hook commands. We render both fresh at install time so the
- * version pin baked into the hook command always matches the installed
- * package. No `goose configure`, config edit, or MCP registration is needed
- * for the gate — discovery is purely directory-based.
+ * describing the hook commands. We render both fresh at install time because
+ * the hook command points at the runtime shim resolved by this install. No
+ * `goose configure`, config edit, or MCP registration is needed for the gate —
+ * discovery is purely directory-based.
  *
  * The plugin dir layout (plugin.json + hooks/hooks.json) and discovery paths
  * are verified against goose v1.39.0.
@@ -64,8 +64,8 @@ const path = __importStar(require("node:path"));
 const argus_1 = require("@ory/argus");
 const assets_js_1 = require("./assets.js");
 const PACKAGE_NAME = "@ory/goose";
-const HOOK_BIN = "ory-goose-hook";
 const PLUGIN_NAME = "ory";
+const PACKAGE_ROOT = path.resolve(__dirname, "..", "..");
 /**
  * Resolve the Goose plugin root: `~/.agents/plugins/ory` with --global, else
  * `<projectDir>/.agents/plugins/ory`.
@@ -97,19 +97,18 @@ function renderManifest(version) {
     return {
         name: PLUGIN_NAME,
         version,
-        description: "Ory authentication, authorization, and distributed tracing for Goose",
+        description: "Ory authentication, authorization, and activity auditing for Goose",
     };
 }
 /**
- * Render hooks/hooks.json. Each event maps to the version-pinned hook binary,
- * resolved via npx so it needn't be on PATH after install. PreToolUse is the
- * only event that can block; the rest are observability hooks.
+ * Render hooks/hooks.json. Each event maps to the runtime shim resolved at
+ * install time, so no package resolution happens per tool call. PreToolUse is
+ * the only event that can block; the rest are observability hooks.
  *
  * The hooks.json schema (per-event arrays of `{ type, command }`, optional
  * regex `matcher`/`pattern`) is verified against goose v1.39.0.
  */
-function renderHooks(version) {
-    const command = (0, argus_1.pinHookToLocalRegistry)(`npx -y -p ${PACKAGE_NAME}@${version} ${HOOK_BIN}`);
+function renderHooks(command) {
     const entry = [{ type: "command", command }];
     return {
         hooks: {
@@ -127,20 +126,22 @@ function renderHooks(version) {
         },
     };
 }
-function writePluginTree(pluginRoot, version) {
+function writePluginTree(pluginRoot, version, runtime) {
     fs.mkdirSync(pluginRoot, { recursive: true });
     fs.writeFileSync(path.join(pluginRoot, "plugin.json"), JSON.stringify(renderManifest(version), null, 2) + "\n");
     const hooksDir = path.join(pluginRoot, "hooks");
     fs.mkdirSync(hooksDir, { recursive: true });
-    fs.writeFileSync(path.join(hooksDir, "hooks.json"), JSON.stringify(renderHooks(version), null, 2) + "\n");
+    fs.writeFileSync(path.join(hooksDir, "hooks.json"), JSON.stringify(renderHooks((0, argus_1.requireHookCommand)(runtime)), null, 2) + "\n");
     (0, assets_js_1.installGooseOryAssets)(pluginRoot);
 }
 function runSetup(args) {
     const version = readPackageVersion();
     if (args.print) {
+        // Nothing is resolved or written on a --print run, so the hook command is
+        // shown as the shim path it will point at.
         console.log(JSON.stringify({
             "plugin.json": renderManifest(version),
-            "hooks/hooks.json": renderHooks(version),
+            "hooks/hooks.json": renderHooks(`node "<runtime shim>"`),
         }, null, 2));
         return;
     }
@@ -153,9 +154,26 @@ function runSetup(args) {
         (0, assets_js_1.uninstallGooseOryAssets)(pluginRoot);
         fs.rmSync(pluginRoot, { recursive: true, force: true });
         console.log(`Removed Ory plugin from ${pluginRoot}`);
+        (0, argus_1.removeRuntimeWiring)("goose");
+        for (const pruned of (0, argus_1.pruneRuntimeStores)()) {
+            console.log(`  Removed runtime: ${pruned}`);
+        }
         return;
     }
-    writePluginTree(pluginRoot, version);
+    const runtime = (0, argus_1.wireRuntime)({
+        harness: "goose",
+        packageName: PACKAGE_NAME,
+        packageRoot: PACKAGE_ROOT,
+        installCommand: "npx -y -p @ory/goose ory-goose install",
+        args: process.argv.slice(2),
+    });
+    console.log(runtime.target.kind === "linked"
+        ? `Runtime: linked to ${runtime.target.packageDir} (dev)`
+        : `Runtime: ${runtime.target.packageName}@${runtime.target.version} in ${runtime.target.storeDir}`);
+    for (const pruned of runtime.prunedStores) {
+        console.log(`  Removed stale runtime: ${pruned}`);
+    }
+    writePluginTree(pluginRoot, version, runtime);
     console.log(`Ory plugin installed to ${pluginRoot}`);
     console.log(`  plugin.json + hooks/hooks.json + skills/ written.`);
     (0, argus_1.printNextSteps)("Goose", "npx -y -p @ory/goose ory-goose-setup --uninstall", {
@@ -176,7 +194,10 @@ if (require.main === module) {
     // Direct `-setup --uninstall` path: also clear stored Ory credentials.
     // (The `ory-goose uninstall` command handles this itself.)
     if (process.argv.includes("--uninstall")) {
-        (0, argus_1.clearCredentialsForUninstall)().then(() => process.exit(0), (err) => {
+        (0, argus_1.clearCredentialsForUninstall)({
+            harness: "goose",
+            purge: process.argv.includes("--purge"),
+        }).then(() => process.exit(0), (err) => {
             console.error(err.message ?? err);
             process.exit(1);
         });

package/dist/handlers.d.ts

@@ -1,14 +1,8 @@
-import { OryAgentClient, ensureUserAuthenticated, ensureAgentIdentity } from "@ory/argus";
+import { OryAgentClient, ensureAgentIdentity, ensureUserAuthenticated } from "@ory/argus";
 import type { GooseHookInput, GooseHookOutput } from "./types.js";
 export interface HandleHookEventDeps {
-    /** Test injection point for the user login flow. */
     userLogin?: typeof ensureUserAuthenticated;
-    /** Test injection point for the agent identity gate. */
     agentGate?: typeof ensureAgentIdentity;
 }
-/**
- * Route a Goose hook event to the appropriate Ory integration.
- *
- * Event-name set verified against goose v1.39.0.
- */
+/** Route a Goose hook event to the appropriate Ory integration. */
 export declare function handleHookEvent(input: GooseHookInput, client: OryAgentClient, deps?: HandleHookEventDeps): Promise<GooseHookOutput>;

package/dist/handlers.js

@@ -2,11 +2,8 @@
 Object.defineProperty(exports, "__esModule", { value: true });
 exports.handleHookEvent = handleHookEvent;
 const argus_1 = require("@ory/argus");
-/**
- * Route a Goose hook event to the appropriate Ory integration.
- *
- * Event-name set verified against goose v1.39.0.
- */
+const HARNESS = "goose";
+/** Route a Goose hook event to the appropriate Ory integration. */
 async function handleHookEvent(input, client, deps = {}) {
     const event = input.event;
     client.logger.debug("hook.received", {
@@ -14,216 +11,89 @@ async function handleHookEvent(input, client, deps = {}) {
         sessionId: input.session_id,
         toolName: input.tool_name,
     });
-    // Set trace context so all spans within this hook invocation correlate.
-    client.tracer.setContext({
-        traceId: (0, argus_1.deriveTraceId)(input.session_id),
-        sessionId: input.session_id,
-    });
-    try {
+    return (0, argus_1.withHookContext)(client, { sessionId: input.session_id }, async () => {
         switch (event) {
             case "SessionStart":
-                return await handleSessionStart(input, client, deps);
+                client.logger.info("lifecycle.session_start", {
+                    sessionId: input.session_id,
+                    cwd: input.working_dir,
+                });
+                await (0, argus_1.sessionStart)(client, { harness: HARNESS, ...deps });
+                return {};
+            case "PreToolUse": {
+                const toolName = input.tool_name ?? "unknown";
+                client.logger.info("lifecycle.pre_tool_use", {
+                    sessionId: input.session_id,
+                    toolName,
+                    toolInput: input.tool_input,
+                });
+                const result = await (0, argus_1.gate)(client, {
+                    harness: HARNESS,
+                    toolName,
+                    toolArgs: input.tool_input,
+                    subjectFallback: `session:${input.session_id}`,
+                });
+                return result.blocked
+                    ? { action: "block", message: result.denialMessage }
+                    : {};
+            }
+            case "PostToolUse": {
+                const toolName = input.tool_name ?? "unknown";
+                client.logger.info("lifecycle.post_tool_use", {
+                    sessionId: input.session_id,
+                    toolName,
+                    error: input.error,
+                });
+                (0, argus_1.complete)(client, {
+                    toolName,
+                    input: input.tool_input,
+                    output: input.tool_output,
+                });
+                return {};
+            }
+            case "PostToolUseFailure": {
+                const toolName = input.tool_name ?? "unknown";
+                client.logger.info("lifecycle.post_tool_use", {
+                    sessionId: input.session_id,
+                    toolName,
+                    error: input.error,
+                });
+                (0, argus_1.fail)(client, { toolName, input: input.tool_input, error: input.error });
+                return {};
+            }
             case "SessionEnd":
             case "Stop":
-                return await handleSessionEnd(input, client);
-            case "PreToolUse":
-                return await handlePreToolUse(input, client);
-            case "PostToolUse":
-            case "PostToolUseFailure":
-                return await handlePostToolUse(input, client);
+                client.logger.info("lifecycle.session_end", {
+                    sessionId: input.session_id,
+                    event,
+                });
+                client.logger.activity("session.end", "ok", { attributes: { event } });
+                return {};
             case "UserPromptSubmit":
-                return await handleUserPromptSubmit(input, client);
+                client.logger.info("lifecycle.user_prompt_submit", { sessionId: input.session_id });
+                client.logger.activity("user.prompt", "ok", { attributes: { event } });
+                return {};
             case "BeforeReadFile":
             case "AfterFileEdit":
             case "BeforeShellExecution":
             case "AfterShellExecution":
-                return await handleTraceOnly(input, client);
+                client.logger.info("lifecycle.observe", {
+                    sessionId: input.session_id,
+                    event,
+                    filePath: input.file_path,
+                });
+                client.logger.activity("notification", "ok", {
+                    attributes: {
+                        event,
+                        ...(input.file_path ? { filePath: input.file_path } : {}),
+                        ...(input.command ? { hasCommand: true } : {}),
+                    },
+                });
+                return {};
             default:
                 client.logger.debug("hook.passthrough", { event });
-                client.tracer.record("hook.passthrough", "skipped", {
-                    attributes: { event },
-                });
+                client.logger.activity("hook.passthrough", "skipped", { attributes: { event } });
                 return {};
         }
-    }
-    finally {
-        client.tracer.clearContext();
-    }
-}
-// ─── SessionStart ───────────────────────────────────────────────────
-async function handleSessionStart(input, client, deps) {
-    client.logger.info("lifecycle.session_start", {
-        sessionId: input.session_id,
-        cwd: input.working_dir,
-    });
-    client.tracer.record("session.start", "ok", {});
-    const userGate = deps.userLogin ?? argus_1.ensureUserAuthenticated;
-    // The user login runs every session and never blocks; tool-call
-    // enforcement is governed solely by permissionMode.
-    await userGate(client, {
-        binName: "ory-goose",
-        harness: "goose",
-    });
-    // Resolve the agent identity (machine credentials). Never blocks;
-    // attaches the agent's bearer token to outgoing Ory API calls.
-    const agentGate = deps.agentGate ?? argus_1.ensureAgentIdentity;
-    await agentGate(client, { projectUrl: (0, argus_1.resolveConfig)().projectUrl, harness: "goose" });
-    // User → agent delegation tuple (best-effort, audit-trail only). Written
-    // at most once per install.
-    await (0, argus_1.writeUserDelegatesAgent)(client);
-    return {};
-}
-// ─── PreToolUse ────────────────────────────────────────────────────
-async function handlePreToolUse(input, client) {
-    const toolName = input.tool_name ?? "unknown";
-    client.logger.info("lifecycle.pre_tool_use", {
-        sessionId: input.session_id,
-        toolName,
-        toolInput: input.tool_input,
-    });
-    const inputSummary = (0, argus_1.summarizeToolInput)(toolName, input.tool_input);
-    // In audit-only mode, log the invocation but skip permission checks.
-    if ((0, argus_1.resolveConfig)().auditOnly) {
-        client.tracer.record("tool.invoke", "ok", {
-            attributes: { toolName, ...inputSummary },
-        });
-        return {};
-    }
-    const subject = (0, argus_1.resolveUserSubject)(client, `session:${input.session_id}`);
-    const subjectId = (0, argus_1.subjectLabel)(subject);
-    const namespace = resolveNamespace();
-    try {
-        const outcome = await (0, argus_1.gateToolCall)(client, {
-            harness: "goose",
-            toolName,
-            check: { namespace, object: toolName, relation: "use", ...subject },
-            spanAttributes: { toolName },
-        });
-        // Interactive tools (operator-extensible via ORY_INTERACTIVE_TOOLS):
-        // user.interaction span is already recorded; pass through.
-        if (outcome.kind === "interactive") {
-            return {};
-        }
-        const decision = outcome;
-        if (decision.kind === "fail_open") {
-            return handlePermissionError(decision.error, toolName, client);
-        }
-        const attrs = { toolName, ...inputSummary };
-        const decisionAttrs = decision.spanAttributes;
-        if (decision.kind === "allow") {
-            client.tracer.record("tool.invoke", "ok", {
-                attributes: { ...attrs, ...decisionAttrs, allowed: true },
-            });
-            return {};
-        }
-        if (decision.kind === "observe") {
-            client.tracer.record("tool.block", "denied", {
-                attributes: { ...attrs, ...decisionAttrs, allowed: false, ...(0, argus_1.alertAttributes)(false) },
-            });
-            client.tracer.record("tool.invoke", "ok", {
-                attributes: { ...attrs, ...decisionAttrs, allowed: false, observed: true },
-            });
-            return {};
-        }
-        client.tracer.record("tool.block", "denied", {
-            attributes: { ...attrs, ...decisionAttrs, allowed: false, ...(0, argus_1.alertAttributes)(true) },
-        });
-        return {
-            action: "block",
-            message: (0, argus_1.formatDenialMessage)({ tool: toolName, subjectId, namespace }),
-        };
-    }
-    catch (err) {
-        return handlePermissionError(err, toolName, client);
-    }
-}
-// ─── PostToolUse / PostToolUseFailure ──────────────────────────────
-async function handlePostToolUse(input, client) {
-    const toolName = input.tool_name ?? "unknown";
-    client.logger.info("lifecycle.post_tool_use", {
-        sessionId: input.session_id,
-        toolName,
-        error: input.error,
-    });
-    const status = input.event === "PostToolUseFailure" ? "error" : "ok";
-    client.tracer.record("tool.complete", status, {
-        attributes: {
-            toolName,
-            ...(0, argus_1.summarizeToolInput)(toolName, input.tool_input),
-            ...(0, argus_1.summarizeToolOutput)(toolName, input.tool_output),
-            ...(input.error ? { error: input.error } : {}),
-        },
-    });
-    return {};
-}
-// ─── SessionEnd / Stop ─────────────────────────────────────────────
-async function handleSessionEnd(input, client) {
-    client.logger.info("lifecycle.session_end", {
-        sessionId: input.session_id,
-        event: input.event,
-    });
-    client.tracer.record("session.end", "ok", {
-        attributes: { event: input.event },
-    });
-    return {};
-}
-// ─── UserPromptSubmit ──────────────────────────────────────────────
-//
-// User-facing phase. The human is the decision-maker; record an audit
-// span only and never block.
-async function handleUserPromptSubmit(input, client) {
-    client.logger.info("lifecycle.user_prompt_submit", {
-        sessionId: input.session_id,
-    });
-    client.tracer.record("user.prompt", "ok", {
-        attributes: { event: input.event },
-    });
-    return {};
-}
-// ─── Trace-only file / shell observation events ────────────────────
-//
-// BeforeReadFile, AfterFileEdit, BeforeShellExecution, AfterShellExecution
-// are fine-grained observability hooks. The PreToolUse gate already
-// authorizes the tool that drives them, so these are audit spans only.
-async function handleTraceOnly(input, client) {
-    client.logger.info("lifecycle.observe", {
-        sessionId: input.session_id,
-        event: input.event,
-        filePath: input.file_path,
-    });
-    client.tracer.record("notification", "ok", {
-        attributes: {
-            event: input.event,
-            ...(input.file_path ? { filePath: input.file_path } : {}),
-            ...(input.command ? { hasCommand: true } : {}),
-        },
-    });
-    return {};
-}
-// ─── Helpers ────────────────────────────────────────────────────────
-function resolveNamespace() {
-    return process.env.ORY_PERMISSION_NAMESPACE ?? "AgentTools";
-}
-function handlePermissionError(oryErr, toolName, client) {
-    if (oryErr.code === "network_error") {
-        client.logger.warn("permission.network_error", {
-            toolName,
-            message: "Ory unreachable, failing open",
-        });
-        return {};
-    }
-    if (oryErr.code === "rate_limited") {
-        client.logger.warn("permission.rate_limited", {
-            toolName,
-            message: "Ory rate limited, failing open",
-        });
-        return {};
-    }
-    // For other errors, also fail open but log prominently.
-    client.logger.error("permission.check.error", {
-        toolName,
-        code: oryErr.code,
-        message: oryErr.message,
     });
-    return {};
 }

package/dist/hook.js

@@ -13,6 +13,16 @@
 Object.defineProperty(exports, "__esModule", { value: true });
 const argus_1 = require("@ory/argus");
 const handlers_js_1 = require("./handlers.js");
+const EVENT_FLUSH_TIMEOUT_MS = 750;
+let client;
+async function flushEvents() {
+    try {
+        await client?.flushEvents(EVENT_FLUSH_TIMEOUT_MS);
+    }
+    catch {
+        // The hook decision must not depend on activity delivery.
+    }
+}
 /**
  * Read all of stdin as a string.
  *
@@ -46,7 +56,7 @@ function readStdin() {
     });
 }
 async function main() {
-    const client = argus_1.OryAgentClient.fromEnv("goose");
+    client = argus_1.OryAgentClient.fromEnv("goose");
     const raw = await readStdin();
     let input;
     try {
@@ -54,7 +64,6 @@ async function main() {
     }
     catch {
         client.logger.error("hook.stdin.parse_failed", { raw: raw.slice(0, 200) });
-        await client.tracer.shutdown();
         process.exit(0); // Don't block on parse errors
     }
     const output = await (0, handlers_js_1.handleHookEvent)(input, client);
@@ -63,13 +72,14 @@ async function main() {
         // signal is the `action` key; emit it on stdout and exit 2 as a defensive
         // secondary signal — both indicate a hard deny.
         process.stdout.write(JSON.stringify(output));
-        await client.tracer.shutdown();
+        await flushEvents();
         process.exit(2);
     }
-    await client.tracer.shutdown();
+    await flushEvents();
     process.exit(0);
 }
-main().catch((err) => {
+main().catch(async (err) => {
     process.stderr.write(`[ory-agent] fatal: ${err}\n`);
+    await flushEvents();
     process.exit(0); // Fail open
 });

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@ory/goose",
-  "version": "0.14.0",
+  "version": "1.0.0",
   "description": "Ory plugin for Goose (Block's open-source agent CLI): scaffolding skills, a local Ory instance, and authentication, authorization, and audit for every tool call",
   "license": "Apache-2.0",
   "homepage": "https://ory.com",
@@ -35,8 +35,7 @@
     "agent-security",
     "guardrails",
     "llm",
-    "tracing",
-    "distributed-tracing",
+    "activity-auditing",
     "observability",
     "kratos",
     "keto",
@@ -47,6 +46,10 @@
     "registry": "https://registry.npmjs.org/",
     "provenance": true
   },
+  "reova": {
+    "enabled": true,
+    "endpoint": "https://telemetry.reo.dev/data"
+  },
   "main": "dist/index.js",
   "types": "dist/index.d.ts",
   "exports": {
@@ -62,12 +65,11 @@
   },
   "files": [
     "dist",
-    "!dist/dev",
     "!dist/**/*.tsbuildinfo"
   ],
   "dependencies": {
-    "reo-census": "^1.2.8",
-    "@ory/argus": "0.14.0"
+    "reova": "^0.7.0",
+    "@ory/argus": "1.0.0"
   },
   "devDependencies": {
     "typescript": "^6.0.2",
@@ -81,7 +83,6 @@
     "clean": "rm -rf dist *.tsbuildinfo",
     "test": "vitest run",
     "test:watch": "vitest",
-    "typecheck": "tsc --noEmit",
-    "dev": "node dist/dev/launcher.js"
+    "typecheck": "tsc --noEmit"
   }
 }