[patch] agent-hooks v0.12.2: Windows hooks plus APR docs and plugin guide ports

.codex-plugin/plugin.json

@@ -1,6 +1,6 @@
 {
   "name": "jfrog",
-  "version": "0.1.10",
+  "version": "0.1.11",
   "description": "JFrog skills and the JFrog MCP server for Codex \u2014 interact with the JFrog Platform.",
   "author": {
     "name": "JFrog",

.github/scripts/sync-modules-vendor.json

@@ -1,6 +1,6 @@
 {
   "repo": "JFROG/jfrog-agent-hooks",
-  "pin": "jfrog-agent-hooks/v0.12.1",
+  "pin": "jfrog-agent-hooks/v0.12.2",
   "paths": [
     "modules"
   ]

docs/package-resolution-admin-guide.md

@@ -4,22 +4,21 @@ Route AI-assisted package installs through your JFrog Artifactory repositories w
 
 Agent Package Resolution runs at the start of each agent session. When enabled, it injects routing policy and resolved Artifactory URLs into the session so the agent prefers your repositories over public registries. Durable enforcement still comes from **package manager configuration** (`jf setup`) and **JFrog Curation** on the server.
 
-This guide is for **platform administrators** and **developers** onboarding the JFrog coding-agent plugins. For installing the plugin itself, see the JFrog documentation for your IDE ([Cursor](https://docs.jfrog.com/ai-ml/docs/cursor), [Claude Code](https://docs.jfrog.com/ai-ml/docs/claude-code/), [VS Code](https://docs.jfrog.com/ai-ml/docs/vs-code)) or this repository's [README](../README.md) for **Codex**.
+This guide is for **platform administrators** and **developers** onboarding the JFrog coding-agent plugins. For installing the plugin itself, see the JFrog documentation for your IDE ([Cursor](https://docs.jfrog.com/ai-ml/docs/cursor), [Claude Code](https://docs.jfrog.com/ai-ml/docs/claude-code/), [VS Code](https://docs.jfrog.com/ai-ml/docs/vs-code)) or the [Codex plugin README](https://github.com/jfrog/codex-plugin#installation) for Codex.
 
 > **Related:** [Use the MCP Registry with Agent Guard](https://docs.jfrog.com/ai-ml/docs/configure-coding-agents) covers MCP governance. Agent Package Resolution is a separate capability in the same JFrog plugin family and uses the same local configuration file for admin settings.
 
 ---
 
 ## Setup summary
 
-| Step | Action                                                                                                     |
-| ---- | ---------------------------------------------------------------------------------------------------------- |
-| 1    | Install the JFrog plugin in your coding assistant                                                          |
-| 2    | Install and configure the JFrog CLI (`jf config add`) — required for **routing** mode                      |
-| 3    | Confirm `~/.jfrog/agents-conf.json` (shipped template enables APR with empty bindings; or deploy your own) |
-| 4    | **Codex:** restart, then `/hooks` and trust the SessionStart command (plugin install does not skip this)   |
-| 5    | Start a **new agent session** — policy and URLs are injected once per session                              |
-
+| Step | Action                                                                                                                    |
+| ---- | ------------------------------------------------------------------------------------------------------------------------- |
+| 1    | Install the JFrog plugin in your coding assistant                                                                         |
+| 2    | Install and configure the JFrog CLI (`jf config add`) — required for **routing** mode                                     |
+| 3    | If an administrator pre-deployed `.jfrog/agents-conf.json`, confirm it; otherwise continue—the first session creates it    |
+| 4    | **Codex:** after install, restart, then `/hooks` and trust the SessionStart command (plugin install does not skip this — see the [Codex plugin README](https://github.com/jfrog/codex-plugin#installation)) |
+| 5    | Start a **new agent session** — missing config is created, then policy and URLs are injected once per session               |
 
 ### Codex install and hook trust
 
@@ -35,16 +34,14 @@ After `codex plugin add jfrog@codex-plugin`:
 
 The hook command uses `${PLUGIN_ROOT}` (Codex also sets `CLAUDE_PLUGIN_ROOT` for compatibility). Local `make install-codex` from `jfrog-agent-hooks` writes a different command into `~/.codex/hooks.json`; that trust hash does **not** cover the plugin definition. Uninstall the local hook before smoke-testing the plugin so you are not approving two SessionStart entries.
 
-The shipped template turns Agent Package Resolution **on** (`enabled: true`) with empty `defaultGlobalRepos`. Nothing is routed until Consent Enable or an administrator adds bindings. Set `enabled: false` or `JF_AGENT_PACKAGE_RESOLUTION_DISABLE=1` to keep it off.
-
+The shipped template turns Agent Package Resolution **on** (`enabled: true`) with empty `defaultGlobalRepos`. Nothing is routed until Consent Enable or an administrator adds bindings. `JF_AGENT_PACKAGE_RESOLUTION_DISABLE=1` always keeps it off; `enabled: false` alone only keeps it off if you deploy your own `agents-conf.json` — on the plugin's auto-created default file it is not durable (see below).
 
 **At a glance:**
 
 - **Default:** on, but routes nothing until you add repositories.
 - **To route installs:** add repository keys under `defaultGlobalRepos` (org config or Consent Enable).
 - **To turn it off org-wide:** deploy your own `agents-conf.json` with `"enabled": false` (see [Turning Agent Package Resolution off](#turning-agent-package-resolution-off-admins)). Setting `"enabled": false` on the plugin's **default file without also deploying your own** is not durable — the plugin re-enables it on the next session.
 
-
 ---
 
 ## Prerequisites
@@ -151,9 +148,6 @@ the `jfrog` skill + `verify-repo`, or manually — see
 With default `verifyRepos: true`, Consent Enable / `configure.mjs enable` accepts
 keys Artifactory confirms as virtual repositories of the requested package type.
 
----
-
-
 ### Turning Agent Package Resolution off (admins)
 
 > **Why `enabled: false` alone may not stick.** Because the feature now ships **on**, the plugin re-enables its **own default file** if it finds it still turned off. "Default file" means the `agents-conf.json` the plugin auto-created and that no one has changed except (at most) the `enabled` flag. As soon as you deploy your **own** config, or add any other setting (like `onboardingPrompt`), the plugin treats it as yours and never re-enables it.
@@ -179,6 +173,8 @@ keys Artifactory confirms as virtual repositories of the requested package type.
 
 Setting **only** `"onboardingPrompt": "off"` stops the Consent Enable prompts but does **not** turn the feature off — leave `enabled: false` in place for that. See also [emergency disable](#environment-variable-emergency-disable) for the environment variable.
 
+---
+
 ## Admin control: deploy `agents-conf.json` across your organization
 
 Use standard endpoint management to place a consistent `agents-conf.json` on every developer machine.
@@ -200,8 +196,6 @@ Use standard endpoint management to place a consistent `agents-conf.json` on eve
 | Auto-configure package managers at first session | Add types to `autoSetup` ([Zero-touch setup](#zero-touch-setup-autosetup))                                                                                   |
 | Force all cached state to refresh                | Set `"cacheTtlDays": 0` (this also re-runs eligible zero-touch `jf setup` each session), or edit `agents-conf.json`                                          |
 | Support troubleshooting                          | Set `"logLevel": "debug"` temporarily; logs go to `~/.jfrog/logs/agent-hooks.log`                                                                            |
-| Keep APR **off** (durable)                       | Deploy your own file with `"enabled": false`, **or** set `"enabled": false` **and** `"onboardingPrompt": "off"` on the plugin's default file — see [Turning off](#turning-agent-package-resolution-off-admins) |
-| Silence Consent Enable offers only               | Set `"onboardingPrompt": "off"` (does not disable APR while `enabled` is `true`)                                                                             |
 
 ---
 
@@ -529,5 +523,5 @@ Workspace values win over `agents-conf.json` for matching types during that sess
 - [Install JFrog Plugin for Cursor](https://docs.jfrog.com/ai-ml/docs/install-jfrog-plugin-for-cursor)
 - [Install JFrog Plugin for Claude Code](https://docs.jfrog.com/ai-ml/docs/install-jfrog-plugin-for-claude-code)
 - [Install JFrog Plugin for VS Code](https://docs.jfrog.com/ai-ml/docs/install-jfrog-plugin-for-vs-code)
-- [Install the JFrog plugin for Codex](../README.md#installation)
+- [Install JFrog Plugin for Codex](https://github.com/jfrog/codex-plugin#installation)
 - [Use the MCP Registry with Agent Guard](https://docs.jfrog.com/ai-ml/docs/configure-coding-agents)

modules/core/agent-guard-check.mjs

@@ -17,6 +17,7 @@ import { execFileSync } from "node:child_process";
 import process from "node:process";
 
 import { isMainEntry } from "./entry.mjs";
+import { jfInvocation } from "./jf-process.mjs";
 import { skillsProductUserAgent } from "./jf-user-agent.mjs";
 
 export const SETTINGS_PATH =
@@ -88,6 +89,7 @@ export function resolveAgentGuardCredentials(opts = {}) {
   if (explicitServerId) {
     const fromCli = resolveFromCliConfig({
       serverId: explicitServerId,
+      env,
       execFileSyncFn: execFn,
       debug,
     });
@@ -114,6 +116,7 @@ export function resolveAgentGuardCredentials(opts = {}) {
   if (explicitServerId) return null;
   return resolveFromCliConfig({
     serverId: undefined,
+    env,
     execFileSyncFn: execFn,
     debug,
   });
@@ -122,22 +125,28 @@ export function resolveAgentGuardCredentials(opts = {}) {
 /**
  * @param {{
  *   serverId?: string,
+ *   env?: NodeJS.ProcessEnv,
  *   execFileSyncFn?: typeof execFileSync,
  *   debug?: (message: string) => void,
  * }} opts
  */
 function resolveFromCliConfig(opts) {
   const debug = opts.debug ?? (() => {});
+  const env = opts.env ?? process.env;
   const execFn = opts.execFileSyncFn ?? execFileSync;
   const exportArgs = opts.serverId
     ? ["config", "export", opts.serverId]
     : ["config", "export"];
+  const invocation = jfInvocation(exportArgs, { env });
   let exported;
   try {
-    exported = execFn("jf", exportArgs, {
+    exported = execFn(invocation.command, invocation.args, {
       encoding: "utf8",
       stdio: ["ignore", "pipe", "ignore"],
       timeout: 2000,
+      windowsHide: true,
+      shell: false,
+      env,
     }).trim();
   } catch (error) {
     debug(

modules/core/jf-identity.mjs

@@ -20,10 +20,10 @@
 // async.
 
 import { createHash } from "node:crypto";
-import { spawnSync } from "node:child_process";
 import process from "node:process";
 
 import { createLogger } from "./logger.mjs";
+import { spawnJfSync } from "./jf-process.mjs";
 import { skillsProductUserAgent } from "./jf-user-agent.mjs";
 
 const log = createLogger("jf-identity");
@@ -82,7 +82,7 @@ function jfConfigIdentity(serverId) {
 
   let result;
   try {
-    result = spawnSync("jf", args, {
+    result = spawnJfSync(args, {
       encoding: "utf8",
       timeout: 2000,
       stdio: ["ignore", "pipe", "pipe"],

modules/core/jf-process.mjs

@@ -0,0 +1,50 @@
+// Shell-free JFrog CLI process boundary shared by coding-agent hooks.
+//
+// Production invokes the native CLI directly (`jf.exe` on Windows, `jf`
+// elsewhere). Tests may substitute the Node fake only when the explicit test
+// harness gate is active; normal production environments ignore the driver.
+
+import { spawnSync } from "node:child_process";
+import process from "node:process";
+
+/**
+ * Resolve the shell-free executable and argv for a JFrog CLI invocation.
+ * @param {string[]} args
+ * @param {{ env?: NodeJS.ProcessEnv, platform?: NodeJS.Platform }} [options]
+ * @returns {{ command: string, args: string[] }}
+ */
+export function jfInvocation(
+  args,
+  { env = process.env, platform = process.platform } = {},
+) {
+  const testDriver =
+    env.JFROG_TEST_HARNESS === "1" ? env.JFROG_TEST_JF_DRIVER : undefined;
+  return testDriver
+    ? { command: process.execPath, args: [testDriver, ...args] }
+    : { command: platform === "win32" ? "jf.exe" : "jf", args };
+}
+
+/**
+ * Synchronously invoke the JFrog CLI without a shell.
+ *
+ * Test-only injection requires both:
+ *   JFROG_TEST_HARNESS=1
+ *   JFROG_TEST_JF_DRIVER=/absolute/path/to/fake-jf.mjs
+ *
+ * @param {string[]} args
+ * @param {import("node:child_process").SpawnSyncOptionsWithStringEncoding} [options]
+ * @returns {import("node:child_process").SpawnSyncReturns<string>}
+ */
+export function spawnJfSync(args, options = {}) {
+  const env = options.env ?? process.env;
+  const spawnSyncFn = options.spawnSyncFn ?? spawnSync;
+  const rest = { ...options };
+  delete rest.spawnSyncFn;
+  const invocation = jfInvocation(args, { env });
+  return spawnSyncFn(invocation.command, invocation.args, {
+    windowsHide: true,
+    shell: false,
+    ...rest,
+    env,
+  });
+}

modules/core/jf-user-agent.mjs

@@ -15,12 +15,12 @@
 // Spawn env is inherited so CLI DetectExecutionContext can append
 // ai-agent/ / ai-client/ / ai-model/ when those signals exist at jf start.
 
-import { spawnSync } from "node:child_process";
+import { spawnJfSync } from "./jf-process.mjs";
 
 // Plugin sync stamps this literal with the release semver (jfrog-sync-modules.py
 // stamp). Only `modules/` is vendored, so nothing outside this tree is readable
 // at runtime. Unstamped trees (this repo, local dev) report 0.0.0.
-const PKG_VERSION = "0.12.1";
+const PKG_VERSION = "0.12.2";
 
 /** Product UA for plugin `fetch()` (no trigger, no jfrog-cli-go). */
 export function skillsProductUserAgent() {
@@ -55,7 +55,7 @@ function resolveCliVersion(env = process.env) {
   try {
     // Keep process PATH/HOME even when callers pass a sparse env object
     // (unit tests often pass only UA-related keys).
-    const res = spawnSync("jf", ["--version"], {
+    const res = spawnJfSync(["--version"], {
       encoding: "utf8",
       timeout: 3000,
       env: { ...process.env, ...env },

modules/core/rewrite-mcp-json.mjs

@@ -27,6 +27,7 @@ import path from "node:path";
 import process from "node:process";
 
 import { EXIT_ENABLED, runAgentGuardCheck } from "./agent-guard-check.mjs";
+import { jfInvocation } from "./jf-process.mjs";
 import { createLogger } from "./logger.mjs";
 
 const log = createLogger("rewrite-mcp-json");
@@ -285,13 +286,16 @@ export function pickDefaultJfCliServer(servers) {
 export function listJfCliServers(opts = {}) {
   const env = opts.env ?? process.env;
   const spawnSyncFn = opts.spawnSyncFn ?? spawnSync;
+  const invocation = jfInvocation(["config", "show", "--format=json"], { env });
   let res;
   try {
-    res = spawnSyncFn("jf", ["config", "show", "--format=json"], {
+    res = spawnSyncFn(invocation.command, invocation.args, {
       encoding: "utf8",
       timeout: 5_000,
       env,
       stdio: ["ignore", "pipe", "pipe"],
+      windowsHide: true,
+      shell: false,
     });
   } catch {
     return [];

modules/package-resolution/scripts/apr-heartbeat.mjs

@@ -22,6 +22,7 @@ import path from "node:path";
 import { createLogger } from "../../core/logger.mjs";
 import { getPlatformIdentity } from "../../core/jf-identity.mjs";
 import { envWithHookUserAgent } from "../../core/jf-user-agent.mjs";
+import { jfInvocation } from "../../core/jf-process.mjs";
 
 const log = createLogger("apr-heartbeat");
 
@@ -183,12 +184,16 @@ export function tryAcquireHeartbeatLock(serverId, opts = {}) {
  * Detached `jf rt ping --server-id <id>` with hook User-Agent.
  * Waits for spawn success vs async error before unref.
  * @param {string} serverId
- * @param {{ spawn?: typeof spawn, env?: NodeJS.ProcessEnv }} [opts]
+ * @param {{ spawn?: typeof spawn, env?: NodeJS.ProcessEnv, platform?: NodeJS.Platform }} [opts]
  * @returns {Promise<boolean>} true if the process started
  */
 export function spawnHeartbeatPing(serverId, opts = {}) {
   const spawnImpl = opts.spawn ?? spawn;
   const env = opts.env ?? process.env;
+  const invocation = jfInvocation(["rt", "ping", "--server-id", serverId], {
+    env,
+    platform: opts.platform,
+  });
   return new Promise((resolve) => {
     let settled = false;
     const finish = (ok) => {
@@ -198,8 +203,9 @@ export function spawnHeartbeatPing(serverId, opts = {}) {
     };
     let child;
     try {
-      child = spawnImpl("jf", ["rt", "ping", "--server-id", serverId], {
+      child = spawnImpl(invocation.command, invocation.args, {
         detached: true,
+        windowsHide: true,
         stdio: "ignore",
         env: envWithHookUserAgent(env),
       });

modules/package-resolution/scripts/eager-setup.mjs

@@ -59,6 +59,8 @@ import {
   packageManagerBinaryOnPath,
 } from "./package-manager-family.mjs";
 import { detectSetupConflict } from "./setup-conflict.mjs";
+import { isMainEntry } from "../../core/entry.mjs";
+import { spawnJfSync } from "../../core/jf-process.mjs";
 import { envWithHookUserAgent } from "../../core/jf-user-agent.mjs";
 
 const log = createLogger("eager-setup");
@@ -247,6 +249,7 @@ function spawnWorker(payloadB64, jobCount = 1) {
   if (process.env.JFROG_EAGER_SETUP_SYNC === "1") {
     spawnSync(process.execPath, [workerPath(), "--run", payloadB64], {
       stdio: "ignore",
+      windowsHide: true,
       env: process.env,
       timeout: syncWorkerTimeoutMs(jobCount),
     });
@@ -255,6 +258,7 @@ function spawnWorker(payloadB64, jobCount = 1) {
   try {
     const child = spawn(process.execPath, [workerPath(), "--run", payloadB64], {
       detached: true,
+      windowsHide: true,
       stdio: "ignore",
       env: process.env,
     });
@@ -643,7 +647,7 @@ export function releaseLock() {
 function supportedPackageManagers() {
   try {
     // --help is local (no Artifactory traffic); no UA needed for telemetry.
-    const res = spawnSync("jf", ["setup", "--help"], {
+    const res = spawnJfSync(["setup", "--help"], {
       encoding: "utf8",
       timeout: 5000,
       env: process.env,
@@ -701,7 +705,7 @@ function runJfSetup(packageManager, serverId, repoKey) {
     "--repo",
     repoKey,
   ];
-  const res = spawnSync("jf", args, {
+  const res = spawnJfSync(args, {
     encoding: "utf8",
     timeout: PER_PACKAGE_MANAGER_TIMEOUT_MS,
     env: envWithHookUserAgent(process.env),
@@ -838,8 +842,7 @@ export async function runWorker(payload) {
 // CLI entry (worker mode)
 // ---------------------------------------------------------------------------
 
-const isMain = import.meta.url === `file://${process.argv[1]}`;
-if (isMain && process.argv[2] === "--run") {
+if (isMainEntry(import.meta.url) && process.argv[2] === "--run") {
   const b64 = process.argv[3];
   try {
     const payload = JSON.parse(Buffer.from(b64, "base64").toString("utf8"));

modules/package-resolution/templates/package-resolution-unconfigured.md

@@ -27,9 +27,9 @@ Do not delegate governed installs to `agent -p` or another child while routing i
 ## After setup — load the routing policy
 
 Once `jf` is configured and `jfrog-setup-package-managers` has bound package
-managers, load the current routing policy:
+managers, load the current routing policy. Run this command:
 
-```bash
+```text
 {{REFRESH_COMMAND}}
 ```
 

package.json

@@ -1,6 +1,6 @@
 {
   "name": "codex-plugin",
-  "version": "0.1.10",
+  "version": "0.1.11",
   "private": true,
   "type": "module",
   "description": "JFrog skills and MCP server plugin for OpenAI Codex.",