Merge pull request #1237 from guyb1/feat/onecli-integration

feat: replace credential proxy with OneCLI gateway for secret injection
2026-03-24 14:55:41 +02:00
parent deee4b2a96 def3748d02
commit d8cc230227
10 changed files with 150 additions and 79 deletions
--- a/.claude/skills/setup/SKILL.md
+++ b/.claude/skills/setup/SKILL.md
@@ -50,7 +50,7 @@ Already configured. Continue.

 **Verify:** `git remote -v` should show `origin` → user's repo, `upstream` → `qwibitai/nanoclaw.git`.

-## 1. Bootstrap (Node.js + Dependencies)
+## 1. Bootstrap (Node.js + Dependencies + OneCLI)

 Run `bash setup.sh` and parse the status block.

@@ -62,6 +62,34 @@ Run `bash setup.sh` and parse the status block.
 - If NATIVE_OK=false → better-sqlite3 failed to load. Install build tools and re-run.
 - Record PLATFORM and IS_WSL for later steps.

+After bootstrap succeeds, install OneCLI and its CLI tool:
+
+```bash
+curl -fsSL onecli.sh/install | sh
+curl -fsSL onecli.sh/cli/install | sh
+```
+
+Verify both installed: `onecli version`. If the command is not found, the CLI was likely installed to `~/.local/bin/`. Add it to PATH for the current session and persist it:
+
+```bash
+export PATH="$HOME/.local/bin:$PATH"
+# Persist for future sessions (append to shell profile if not already present)
+grep -q '.local/bin' ~/.bashrc 2>/dev/null || echo 'export PATH="$HOME/.local/bin:$PATH"' >> ~/.bashrc
+grep -q '.local/bin' ~/.zshrc 2>/dev/null || echo 'export PATH="$HOME/.local/bin:$PATH"' >> ~/.zshrc
+```
+
+Then re-verify with `onecli version`.
+
+Point the CLI at the local OneCLI instance (it defaults to the cloud service otherwise):
+```bash
+onecli config set api-host http://127.0.0.1:10254
+```
+
+Ensure `.env` has the OneCLI URL (create the file if it doesn't exist):
+```bash
+grep -q 'ONECLI_URL' .env 2>/dev/null || echo 'ONECLI_URL=http://127.0.0.1:10254' >> .env
+```
+
 ## 2. Check Environment

 Run `npx tsx setup/index.ts --step environment` and parse the status block.
@@ -112,15 +140,47 @@ Run `npx tsx setup/index.ts --step container -- --runtime <chosen>` and parse th

 **If TEST_OK=false but BUILD_OK=true:** The image built but won't run. Check logs — common cause is runtime not fully started. Wait a moment and retry the test.

-## 4. Claude Authentication (No Script)
+## 4. Anthropic Credentials via OneCLI

-If HAS_ENV=true from step 2, read `.env` and check for `CLAUDE_CODE_OAUTH_TOKEN` or `ANTHROPIC_API_KEY`. If present, confirm with user: keep or reconfigure?
+NanoClaw uses OneCLI to manage credentials — API keys are never stored in `.env` or exposed to containers. The OneCLI gateway injects them at request time.

-AskUserQuestion: Claude subscription (Pro/Max) vs Anthropic API key?
+Check if a secret already exists:
+```bash
+onecli secrets list
+```

-**Subscription:** Tell user to run `claude setup-token` in another terminal, copy the token, add `CLAUDE_CODE_OAUTH_TOKEN=<token>` to `.env`. Do NOT collect the token in chat.
+If an Anthropic secret is listed, confirm with user: keep or reconfigure? If keeping, skip to step 5.

-**API key:** Tell user to add `ANTHROPIC_API_KEY=<key>` to `.env`.
+AskUserQuestion: Do you want to use your **Claude subscription** (Pro/Max) or an **Anthropic API key**?
+
+1. **Claude subscription (Pro/Max)** — description: "Uses your existing Claude Pro or Max subscription. You'll run `claude setup-token` in another terminal to get your token."
+2. **Anthropic API key** — description: "Pay-per-use API key from console.anthropic.com."
+
+### Subscription path
+
+Tell the user to run `claude setup-token` in another terminal and copy the token it outputs. Do NOT collect the token in chat.
+
+Once they have the token, they register it with OneCLI. AskUserQuestion with two options:
+
+1. **Dashboard** — description: "Best if you have a browser on this machine. Open http://127.0.0.1:10254 and add the secret in the UI. Use type 'anthropic' and paste your token as the value."
+2. **CLI** — description: "Best for remote/headless servers. Run: `onecli secrets create --name Anthropic --type anthropic --value YOUR_TOKEN --host-pattern api.anthropic.com`"
+
+### API key path
+
+Tell the user to get an API key from https://console.anthropic.com/settings/keys if they don't have one.
+
+Then AskUserQuestion with two options:
+
+1. **Dashboard** — description: "Best if you have a browser on this machine. Open http://127.0.0.1:10254 and add the secret in the UI."
+2. **CLI** — description: "Best for remote/headless servers. Run: `onecli secrets create --name Anthropic --type anthropic --value YOUR_KEY --host-pattern api.anthropic.com`"
+
+### After either path
+
+Ask them to let you know when done.
+
+**If the user's response happens to contain a token or key** (starts with `sk-ant-`): handle it gracefully — run the `onecli secrets create` command with that value on their behalf.
+
+**After user confirms:** verify with `onecli secrets list` that an Anthropic secret exists. If not, ask again.

 ## 5. Set Up Channels

@@ -198,7 +258,7 @@ Run `npx tsx setup/index.ts --step verify` and parse the status block.
 **If STATUS=failed, fix each:**
 - SERVICE=stopped → `npm run build`, then restart: `launchctl kickstart -k gui/$(id -u)/com.nanoclaw` (macOS) or `systemctl --user restart nanoclaw` (Linux) or `bash start-nanoclaw.sh` (WSL nohup)
 - SERVICE=not_found → re-run step 7
- CREDENTIALS=missing → re-run step 4
+- CREDENTIALS=missing → re-run step 4 (check `onecli secrets list` for Anthropic secret)
 - CHANNEL_AUTH shows `not_found` for any channel → re-invoke that channel's skill (e.g. `/add-telegram`)
 - REGISTERED_GROUPS=0 → re-invoke the channel skills from step 5
 - MOUNT_ALLOWLIST=missing → `npx tsx setup/index.ts --step mounts -- --empty`
@@ -207,7 +267,7 @@ Tell user to test: send a message in their registered chat. Show: `tail -f logs/

 ## Troubleshooting

-**Service not starting:** Check `logs/nanoclaw.error.log`. Common: wrong Node path (re-run step 7), missing `.env` (step 4), missing channel credentials (re-invoke channel skill).
+**Service not starting:** Check `logs/nanoclaw.error.log`. Common: wrong Node path (re-run step 7), OneCLI not running (check `curl http://127.0.0.1:10254/api/health`), missing channel credentials (re-invoke channel skill).

 **Container agent fails ("Claude Code process exited with code 1"):** Ensure the container runtime is running — `open -a Docker` (macOS Docker), `container system start` (Apple Container), or `sudo systemctl start docker` (Linux). Check container logs in `groups/main/logs/container-*.log`.

--- a/container/Dockerfile
+++ b/container/Dockerfile
@@ -1,7 +1,7 @@
 # NanoClaw Agent Container
 # Runs Claude Agent SDK in isolated Linux VM with browser automation

-FROM node:22-slim
+FROM node:24-slim

 # Install system dependencies for Chromium
 RUN apt-get update && apt-get install -y \
--- a/package-lock.json
+++ b/package-lock.json
@@ -8,6 +8,7 @@
      "name": "nanoclaw",
      "version": "1.2.21",
      "dependencies": {
+        "@onecli-sh/sdk": "^0.2.0",
        "better-sqlite3": "^11.8.1",
        "cron-parser": "^5.5.0",
        "pino": "^9.6.0",
@@ -786,6 +787,15 @@
        "@jridgewell/sourcemap-codec": "^1.4.14"
      }
    },
+    "node_modules/@onecli-sh/sdk": {
+      "version": "0.2.0",
+      "resolved": "https://registry.npmjs.org/@onecli-sh/sdk/-/sdk-0.2.0.tgz",
+      "integrity": "sha512-u7PqWROEvTV9f0ADVkjigTrd2AZn3klbPrv7GGpeRHIJpjAxJUdlWqxr5kiGt6qTDKL8t3nq76xr4X2pxTiyBg==",
+      "license": "MIT",
+      "engines": {
+        "node": ">=20"
+      }
+    },
    "node_modules/@pinojs/redact": {
      "version": "0.4.0",
      "resolved": "https://registry.npmjs.org/@pinojs/redact/-/redact-0.4.0.tgz",
--- a/package.json
+++ b/package.json
@@ -21,6 +21,7 @@
    "test:watch": "vitest"
  },
  "dependencies": {
+    "@onecli-sh/sdk": "^0.2.0",
    "better-sqlite3": "^11.8.1",
    "cron-parser": "^5.5.0",
    "pino": "^9.6.0",
--- a/setup/verify.ts
+++ b/setup/verify.ts
@@ -101,7 +101,7 @@ export async function run(_args: string[]): Promise<void> {
  const envFile = path.join(projectRoot, '.env');
  if (fs.existsSync(envFile)) {
    const envContent = fs.readFileSync(envFile, 'utf-8');
-    if (/^(CLAUDE_CODE_OAUTH_TOKEN|ANTHROPIC_API_KEY)=/m.test(envContent)) {
+    if (/^(CLAUDE_CODE_OAUTH_TOKEN|ANTHROPIC_API_KEY|ONECLI_URL)=/m.test(envContent)) {
      credentials = 'configured';
    }
  }
--- a/src/config.ts
+++ b/src/config.ts
@@ -4,9 +4,7 @@ import path from 'path';
 import { readEnvFile } from './env.js';

 // Read config values from .env (falls back to process.env).
-// Secrets (API keys, tokens) are NOT read here — they are loaded only
-// by the credential proxy (credential-proxy.ts), never exposed to containers.
-const envConfig = readEnvFile(['ASSISTANT_NAME', 'ASSISTANT_HAS_OWN_NUMBER']);
+const envConfig = readEnvFile(['ASSISTANT_NAME', 'ASSISTANT_HAS_OWN_NUMBER', 'ONECLI_URL']);

 export const ASSISTANT_NAME =
  process.env.ASSISTANT_NAME || envConfig.ASSISTANT_NAME || 'Andy';
@@ -47,10 +45,8 @@ export const CONTAINER_MAX_OUTPUT_SIZE = parseInt(
  process.env.CONTAINER_MAX_OUTPUT_SIZE || '10485760',
  10,
 ); // 10MB default
-export const CREDENTIAL_PROXY_PORT = parseInt(
-  process.env.CREDENTIAL_PROXY_PORT || '3001',
-  10,
-);
+export const ONECLI_URL =
+  process.env.ONECLI_URL || envConfig.ONECLI_URL || 'http://localhost:10254';
 export const IPC_POLL_INTERVAL = 1000;
 export const IDLE_TIMEOUT = parseInt(process.env.IDLE_TIMEOUT || '1800000', 10); // 30min default — how long to keep container alive after last result
 export const MAX_CONCURRENT_CONTAINERS = Math.max(
--- a/src/container-runner.test.ts
+++ b/src/container-runner.test.ts
@@ -11,10 +11,10 @@ vi.mock('./config.js', () => ({
  CONTAINER_IMAGE: 'nanoclaw-agent:latest',
  CONTAINER_MAX_OUTPUT_SIZE: 10485760,
  CONTAINER_TIMEOUT: 1800000, // 30min
-  CREDENTIAL_PROXY_PORT: 3001,
  DATA_DIR: '/tmp/nanoclaw-test-data',
  GROUPS_DIR: '/tmp/nanoclaw-test-groups',
  IDLE_TIMEOUT: 1800000, // 30min
+  ONECLI_URL: 'http://localhost:10254',
  TIMEZONE: 'America/Los_Angeles',
 }));

@@ -51,6 +51,15 @@ vi.mock('./mount-security.js', () => ({
  validateAdditionalMounts: vi.fn(() => []),
 }));

+// Mock OneCLI SDK
+vi.mock('@onecli-sh/sdk', () => ({
+  OneCLI: class {
+    applyContainerConfig = vi.fn().mockResolvedValue(true);
+    createAgent = vi.fn().mockResolvedValue({ id: 'test' });
+    ensureAgent = vi.fn().mockResolvedValue({ name: 'test', identifier: 'test', created: true });
+  },
+}));
+
 // Create a controllable fake ChildProcess
 function createFakeProcess() {
  const proc = new EventEmitter() as EventEmitter & {
--- a/src/container-runner.ts
+++ b/src/container-runner.ts
@@ -10,25 +10,26 @@ import {
  CONTAINER_IMAGE,
  CONTAINER_MAX_OUTPUT_SIZE,
  CONTAINER_TIMEOUT,
-  CREDENTIAL_PROXY_PORT,
  DATA_DIR,
  GROUPS_DIR,
  IDLE_TIMEOUT,
+  ONECLI_URL,
  TIMEZONE,
 } from './config.js';
 import { resolveGroupFolderPath, resolveGroupIpcPath } from './group-folder.js';
 import { logger } from './logger.js';
 import {
-  CONTAINER_HOST_GATEWAY,
  CONTAINER_RUNTIME_BIN,
  hostGatewayArgs,
  readonlyMountArgs,
  stopContainer,
 } from './container-runtime.js';
-import { detectAuthMode } from './credential-proxy.js';
+import { OneCLI } from '@onecli-sh/sdk';
 import { validateAdditionalMounts } from './mount-security.js';
 import { RegisteredGroup } from './types.js';

+const onecli = new OneCLI({ url: ONECLI_URL });
+
 // Sentinel markers for robust output parsing (must match agent-runner)
 const OUTPUT_START_MARKER = '---NANOCLAW_OUTPUT_START---';
 const OUTPUT_END_MARKER = '---NANOCLAW_OUTPUT_END---';
@@ -77,7 +78,7 @@ function buildVolumeMounts(
    });

    // Shadow .env so the agent cannot read secrets from the mounted project root.
-    // Credentials are injected by the credential proxy, never exposed to containers.
+    // Credentials are injected by the OneCLI gateway, never exposed to containers.
    const envFile = path.join(projectRoot, '.env');
    if (fs.existsSync(envFile)) {
      mounts.push({
@@ -212,30 +213,26 @@ function buildVolumeMounts(
  return mounts;
 }

-function buildContainerArgs(
+async function buildContainerArgs(
  mounts: VolumeMount[],
  containerName: string,
-): string[] {
+  agentIdentifier?: string,
+): Promise<string[]> {
  const args: string[] = ['run', '-i', '--rm', '--name', containerName];

  // Pass host timezone so container's local time matches the user's
  args.push('-e', `TZ=${TIMEZONE}`);

-  // Route API traffic through the credential proxy (containers never see real secrets)
-  args.push(
-    '-e',
-    `ANTHROPIC_BASE_URL=http://${CONTAINER_HOST_GATEWAY}:${CREDENTIAL_PROXY_PORT}`,
-  );
-
-  // Mirror the host's auth method with a placeholder value.
-  // API key mode: SDK sends x-api-key, proxy replaces with real key.
-  // OAuth mode:   SDK exchanges placeholder token for temp API key,
-  //               proxy injects real OAuth token on that exchange request.
-  const authMode = detectAuthMode();
-  if (authMode === 'api-key') {
-    args.push('-e', 'ANTHROPIC_API_KEY=placeholder');
+  // OneCLI gateway handles credential injection — containers never see real secrets.
+  // The gateway intercepts HTTPS traffic and injects API keys or OAuth tokens.
+  const onecliApplied = await onecli.applyContainerConfig(args, {
+    addHostMapping: false, // Nanoclaw already handles host gateway
+    agent: agentIdentifier,
+  });
+  if (onecliApplied) {
+    logger.info({ containerName }, 'OneCLI gateway config applied');
  } else {
-    args.push('-e', 'CLAUDE_CODE_OAUTH_TOKEN=placeholder');
+    logger.warn({ containerName }, 'OneCLI gateway not reachable — container will have no credentials');
  }

  // Runtime-specific args for host gateway resolution
@@ -278,7 +275,11 @@ export async function runContainerAgent(
  const mounts = buildVolumeMounts(group, input.isMain);
  const safeName = group.folder.replace(/[^a-zA-Z0-9-]/g, '-');
  const containerName = `nanoclaw-${safeName}-${Date.now()}`;
-  const containerArgs = buildContainerArgs(mounts, containerName);
+  // Main group uses the default OneCLI agent; others use their own agent.
+  const agentIdentifier = input.isMain
+    ? undefined
+    : group.folder.toLowerCase().replace(/_/g, '-');
+  const containerArgs = await buildContainerArgs(mounts, containerName, agentIdentifier);

  logger.debug(
    {
--- a/src/container-runtime.ts
+++ b/src/container-runtime.ts
@@ -3,7 +3,6 @@
 * All runtime-specific logic lives here so swapping runtimes means changing one file.
 */
 import { execSync } from 'child_process';
-import fs from 'fs';
 import os from 'os';

 import { logger } from './logger.js';
@@ -11,35 +10,6 @@ import { logger } from './logger.js';
 /** The container runtime binary name. */
 export const CONTAINER_RUNTIME_BIN = 'docker';

-/** Hostname containers use to reach the host machine. */
-export const CONTAINER_HOST_GATEWAY = 'host.docker.internal';
-
-/**
- * Address the credential proxy binds to.
- * Docker Desktop (macOS): 127.0.0.1 — the VM routes host.docker.internal to loopback.
- * Docker (Linux): bind to the docker0 bridge IP so only containers can reach it,
- *   falling back to 0.0.0.0 if the interface isn't found.
- */
-export const PROXY_BIND_HOST =
-  process.env.CREDENTIAL_PROXY_HOST || detectProxyBindHost();
-
-function detectProxyBindHost(): string {
-  if (os.platform() === 'darwin') return '127.0.0.1';
-
-  // WSL uses Docker Desktop (same VM routing as macOS) — loopback is correct.
-  // Check /proc filesystem, not env vars — WSL_DISTRO_NAME isn't set under systemd.
-  if (fs.existsSync('/proc/sys/fs/binfmt_misc/WSLInterop')) return '127.0.0.1';
-
-  // Bare-metal Linux: bind to the docker0 bridge IP instead of 0.0.0.0
-  const ifaces = os.networkInterfaces();
-  const docker0 = ifaces['docker0'];
-  if (docker0) {
-    const ipv4 = docker0.find((a) => a.family === 'IPv4');
-    if (ipv4) return ipv4.address;
-  }
-  return '0.0.0.0';
-}
-
 /** CLI args needed for the container to resolve the host gateway. */
 export function hostGatewayArgs(): string[] {
  // On Linux, host.docker.internal isn't built-in — add it explicitly
--- a/src/index.ts
+++ b/src/index.ts
@@ -1,15 +1,16 @@
 import fs from 'fs';
 import path from 'path';

+import { OneCLI } from '@onecli-sh/sdk';
+
 import {
  ASSISTANT_NAME,
-  CREDENTIAL_PROXY_PORT,
  IDLE_TIMEOUT,
+  ONECLI_URL,
  POLL_INTERVAL,
  TIMEZONE,
  TRIGGER_PATTERN,
 } from './config.js';
-import { startCredentialProxy } from './credential-proxy.js';
 import './channels/index.js';
 import {
  getChannelFactory,
@@ -24,7 +25,6 @@ import {
 import {
  cleanupOrphans,
  ensureContainerRuntimeRunning,
-  PROXY_BIND_HOST,
 } from './container-runtime.js';
 import {
  getAllChats,
@@ -72,6 +72,27 @@ let messageLoopRunning = false;
 const channels: Channel[] = [];
 const queue = new GroupQueue();

+const onecli = new OneCLI({ url: ONECLI_URL });
+
+function ensureOneCLIAgent(jid: string, group: RegisteredGroup): void {
+  if (group.isMain) return;
+  const identifier = group.folder.toLowerCase().replace(/_/g, '-');
+  onecli.ensureAgent({ name: group.name, identifier }).then(
+    (res) => {
+      logger.info(
+        { jid, identifier, created: res.created },
+        'OneCLI agent ensured',
+      );
+    },
+    (err) => {
+      logger.debug(
+        { jid, identifier, err: String(err) },
+        'OneCLI agent ensure skipped',
+      );
+    },
+  );
+}
+
 function loadState(): void {
  lastTimestamp = getRouterState('last_timestamp') || '';
  const agentTs = getRouterState('last_agent_timestamp');
@@ -112,6 +133,9 @@ function registerGroup(jid: string, group: RegisteredGroup): void {
  // Create group folder
  fs.mkdirSync(path.join(groupDir, 'logs'), { recursive: true });

+  // Ensure a corresponding OneCLI agent exists (best-effort, non-blocking)
+  ensureOneCLIAgent(jid, group);
+
  logger.info(
    { jid, name: group.name, folder: group.folder },
    'Group registered',
@@ -474,18 +498,18 @@ async function main(): Promise<void> {
  initDatabase();
  logger.info('Database initialized');
  loadState();
-  restoreRemoteControl();

-  // Start credential proxy (containers route API calls through this)
-  const proxyServer = await startCredentialProxy(
-    CREDENTIAL_PROXY_PORT,
-    PROXY_BIND_HOST,
-  );
+  // Ensure OneCLI agents exist for all registered groups.
+  // Recovers from missed creates (e.g. OneCLI was down at registration time).
+  for (const [jid, group] of Object.entries(registeredGroups)) {
+    ensureOneCLIAgent(jid, group);
+  }
+
+  restoreRemoteControl();

  // Graceful shutdown handlers
  const shutdown = async (signal: string) => {
    logger.info({ signal }, 'Shutdown signal received');
-    proxyServer.close();
    await queue.shutdown(10000);
    for (const ch of channels) await ch.disconnect();
    process.exit(0);