ssh-mcp-pro
<p align="center"><img src="assets/brand/icon-192.png" alt="ssh-mcp-pro icon" width="88" height="88" /></p>
<h1 align="center">ssh-mcp-pro</h1>
<p align="center"><strong>Policy-controlled SSH automation for MCP clients</strong></p>
<p align="center">
<a href="INSTALL.md">Installation</a> ·
<a href="#quickstart">Quickstart</a> ·
<a href="https://oaslananka.github.io/ssh-mcp-pro/">API Docs</a> ·
<a href="https://deepwiki.com/oaslananka/ssh-mcp-pro">DeepWiki</a>
</p>
<p align="center">
<a href="https://www.npmjs.com/package/ssh-mcp-pro"><img alt="npm published version" src="https://img.shields.io/npm/v/ssh-mcp-pro?style=flat&logo=npm" /></a>
<a href="https://github.com/oaslananka/ssh-mcp-pro/releases/latest"><img alt="GitHub release version" src="https://img.shields.io/github/v/release/oaslananka/ssh-mcp-pro?style=flat&label=GitHub%20Release" /></a>
<a href="#prerequisites"><img alt="Node.js supported major versions: 22, 24, 26" src="https://img.shields.io/badge/Node.js-22%20%7C%2024%20%7C%2026-43853d?style=flat&logo=nodedotjs&logoColor=white" /></a>
<a href="LICENSE"><img src="https://img.shields.io/badge/license-AGPL--3.0--only-blue" alt="AGPL-3.0-only license" /></a>
<a href="https://oaslananka.github.io/ssh-mcp-pro/"><img alt="API documentation" src="https://img.shields.io/badge/Docs-API-2980b9?style=flat" /></a>
</p>
<p align="center">
<a href="https://github.com/oaslananka/ssh-mcp-pro/actions/workflows/ci.yml"><img alt="Main CI workflow status" src="https://github.com/oaslananka/ssh-mcp-pro/actions/workflows/ci.yml/badge.svg?branch=main" /></a>
<a href="https://github.com/oaslananka/ssh-mcp-pro/actions/workflows/codeql.yml"><img alt="CodeQL analysis workflow status" src="https://github.com/oaslananka/ssh-mcp-pro/actions/workflows/codeql.yml/badge.svg?branch=main" /></a>
</p>
<p align="center">
<a href="https://www.npmjs.com/package/ssh-mcp-pro"><img src="https://img.shields.io/npm/dt/ssh-mcp-pro?label=total%20downloads" alt="npm total downloads" /></a>
<a href="https://gitgem.org/github/oaslananka/ssh-mcp-pro"><img src="https://gitgem.org/api/badge/github/oaslananka/ssh-mcp-pro.svg" alt="GitGem" /></a>
</p>
<p align="center">
<a href="https://github.com/sponsors/oaslananka"><img src="https://img.shields.io/badge/GitHub-Sponsors-ea4aaa?logo=githubsponsors" alt="GitHub Sponsors" /></a>
<a href="https://www.buymeacoffee.com/oaslananka"><img src="https://img.shields.io/badge/Buy%20Me%20a%20Coffee-Support-ffdd00?logo=buymeacoffee&logoColor=black" alt="Buy Me a Coffee" /></a>
<a href="https://github.com/oaslananka/ssh-mcp-pro"><img src="https://img.shields.io/github/stars/oaslananka/ssh-mcp-pro?label=Stars&color=8e65d3" alt="Stars" /></a>
</p>
ssh-mcp-pro is a secure Model Context Protocol (MCP) server for SSH automation. It lets MCP-capable clients open SSH sessions, inspect hosts, run guarded commands, manage files, transfer artifacts, create tunnels, and perform idempotent package or service work through policy-controlled tools.
## Prerequisites
- Node.js `>=22.22.2` or `>=24.15.0` or `>=26.3.0`
- pnpm `^11.28.11` (use Corepack for development)
- SSH access to the target hosts
- Docker, only for local integration tests and container image builds
## Installation
For Windows/Linux outbound agent connection status, rotating diagnostics and a live PowerShell log view, see [Agent Activity Monitor](docs/agent-activity-monitor.md).
Install globally with pnpm:
```bash
pnpm add --global ssh-mcp-pro
ssh-mcp-pro --version
```
Run without a global install:
```bash
npx ssh-mcp-pro
```
For pnpm-only environments, use:
```bash
pnpm dlx ssh-mcp-pro
```
Container images are published to GitHub Container Registry for release tags:
```bash
docker run --rm ghcr.io/oaslananka/ssh-mcp-pro:1.4.1 --version
```
Images are published for `linux/amd64` and `linux/arm64` with exact semver and
Git tag aliases. Production deployments should prefer the digest-pinned
reference recorded by the release workflow. See [Docker Usage](docs/docker.md)
for the tag policy, digest-pinned examples, and registry verification steps.
## Cloud dashboard and AI connectors
The optional [ssh-mcp-pro Cloud](https://ssh-mcp-pro.oaslananka.dev/) lets you sign in with Google or GitHub, enroll your own devices, and connect an OAuth-capable AI client to `https://ssh-mcp-pro.oaslananka.dev/mcp`. Your SSH private keys remain on your devices.
After installing the package, start a **local web dashboard with one command**:
```bash
ssh-mcp-pro web
```
The shortcut binds only to `http://127.0.0.1:3000/dashboard` and does not publish a privileged server to your network. Signing in requires your own configured GitHub or Google OAuth application. For hosted, multi-user operation use the production control plane configuration; the localhost command is for local development and testing.
The hosted service serves [Privacy](https://ssh-mcp-pro.oaslananka.dev/privacy) and [Terms](https://ssh-mcp-pro.oaslananka.dev/terms). Operators may enable an owner-only **Usage & Users** tab by setting `SSH_MCP_DASHBOARD_OWNER_GITHUB_LOGINS` and/or `SSH_MCP_DASHBOARD_OWNER_GOOGLE_EMAILS`; global usage access is disabled by default. The separately gated [OpenAI reviewer sandbox](docs/openai-reviewer-sandbox.md) uses a dedicated sample-only identity and is not a general sign-in method.
See [Web publication readiness](docs/development/web-publication-readiness.md) for ChatGPT, Claude, and Gemini connection steps and independent acceptance requirements. Manifest validation alone does **not** mean any platform has approved the plugin.
## Quickstart
Generic stdio MCP config:
```json
{
"name": "ssh-mcp-pro",
"command": "ssh-mcp-pro",
"type": "stdio"
}
```
VS Code workspace `.vscode/mcp.json` configuration:
```json
{
"servers": {
"ssh-mcp-pro": {
"type": "stdio",
"command": "ssh-mcp-pro",
"args": []
}
}
}
```
Claude Desktop style:
```json
{
"mcpServers": {
"ssh-mcp-pro": {
"command": "ssh-mcp-pro",
"args": []
}
}
}
```
After registration, start with discovery and a strict host-key policy:
```text
List configured SSH hosts, open a session to bastion.example.com as deploy with hostKeyPolicy=strict, then run os_detect.
```
## Usage
Use ssh-mcp-pro from an MCP client over stdio, or run the HTTP transport for
remote-safe connector profiles. Start with read-only discovery tools, inspect
the active policy, and create explicit sessions before running remote commands:
```text
List configured SSH hosts, explain the active SSH policy, connect to the selected host, then report its operating system and disk usage.
```
See [examples/README.md](examples/README.md) for additional workflows and
[INSTALL.md](INSTALL.md) for client-specific setup.
### General computer control through the hosted connector
Once a user has enrolled a computer, ChatGPT, Claude or another authorized MCP
client can manage it using generic tools, regardless of whether the computer
has Git, Docker, a database, or other applications installed:
- **Run an installed application:** `run_shell` or the managed
`process_start`, `process_read`, `process_input` and `process_stop` tools.
The shell is the native PowerShell on Windows or `sh` on Unix.
- **Find and inspect files:** `file_search`, `file_content_search`,
`file_inspect` (`stat` / `list`) and `file_read`.
- **Manage project files:** `file_manage` (`mkdir`, `copy`, `move`, `remove`),
`file_write`, and the preview/expected-hash `file_replace`.
Directory listings and file searches are bounded. File management operates
on the selected device itself rather than transferring entire files through
the AI conversation.
Choose **Full Admin** for a device when you want unrestricted access *within
that operating-system user account*. New Full Admin profiles can work with
ordinary home, project and mounted-disk paths; the OS account still determines
whether an operation succeeds. The device owner's saved custom allow/deny
rules still take priority. Existing enrolled devices retain their saved
policies until the owner edits them in the dashboard. OAuth scopes separately
control which of those device operations a connected AI client may request.
Dedicated Docker MCP tools are no longer advertised to newly connecting
clients. Previously authorized clients may still use the legacy Docker tool
names during the compatibility period; new users can call an installed
application's own CLI through the generic shell instead. This does **not**
provide GUI screen, mouse, keyboard or browser interaction: those require
desktop-session integration and OS accessibility/screen permissions.
## Configuration
All `SSH_MCP_*` environment variables parsed by `src/config.ts` are listed below. Comma-separated settings also accept newline-separated values.
| Variable | Default | Purpose |
| --- | --- | --- |
| `SSH_MCP_MAX_SESSIONS` | `20` | Maximum concurrent SSH sessions. |
| `SSH_MCP_SESSION_TTL` | `900000` | Session time-to-live in milliseconds. |
| `SSH_MCP_COMMAND_TIMEOUT` | `30000` | Default remote command timeout in milliseconds. |
| `SSH_MCP_MAX_COMMAND_OUTPUT_BYTES` | `1048576` | Maximum buffered stdout/stderr bytes per command result. |
| `SSH_MCP_MAX_STREAM_CHUNKS` | `4096` | Maximum retained streaming chunks. |
| `SSH_MCP_MAX_FILE_SIZE` | `10485760` | Maximum bytes returned by text-focused file reads. |
| `SSH_MCP_MAX_FILE_WRITE_BYTES` | `10485760` | Maximum accepted write payload before buffering. |
| `SSH_MCP_MAX_TRANSFER_BYTES` | `52428800` | Maximum upload or download transfer size. |
| `SSH_MCP_DEBUG` | `false` | Enables debug-oriented configuration behavior. |
| `SSH_MCP_RATE_LIMIT` | `true` | Enables the global MCP request rate limiter. |
| `SSH_MCP_RATE_LIMIT_MAX` | `100` | Maximum requests per rate-limit window. |
| `SSH_MCP_RATE_LIMIT_PER_SESSION` | `true` | Enables per-session MCP request rate limiting when tool arguments include `sessionId`. |
| `SSH_MCP_RATE_LIMIT_PER_SESSION_MAX` | `50` | Maximum requests per SSH session per rate-limit window. |
| `SSH_MCP_RATE_LIMIT_PER_SESSION_WINDOW_MS` | `60000` | Per-session rate-limit window in milliseconds. |
| `SSH_MCP_RATE_LIMIT_WINDOW_MS` | `60000` | Rate-limit window in milliseconds. |
| `SSH_MCP_STRICT_HOST_KEY` | unset | Legacy boolean alias for strict vs insecure host-key checking. |
| `SSH_MCP_HOST_KEY_POLICY` | `strict` | Host-key mode: `strict`, `accept-new`, or `insecure`. |
| `SSH_MCP_KNOWN_HOSTS_PATH` | `~/.ssh/known_hosts` | Known hosts file used for strict host-key verification. |
| `SSH_MCP_ALLOW_ROOT_LOGIN` | `false` | Allows SSH login as root and mirrors into policy. |
| `SSH_MCP_ALLOWED_CIPHERS` | empty | Optional SSH cipher allowlist. |
| `SSH_MCP_POLICY_FILE` | unset | JSON file containing partial policy overrides. |
| `SSH_MCP_POLICY_MODE` | `enforce` | Policy decision mode: `enforce` or `explain`. |
| `SSH_MCP_ALLOW_RAW_SUDO` | `false` | Allows raw `proc_sudo`; prefer `ensure_*` tools. |
| `SSH_MCP_ALLOW_DESTRUCTIVE_COMMANDS` | `false` | Allows commands matching destructive command policy. |
| `SSH_MCP_ALLOW_DESTRUCTIVE_FS` | `false` | Allows destructive filesystem operations such as `fs_rmrf`. |
| `SSH_MCP_ALLOWED_HOSTS` | empty | Host allowlist for policy and remote connector safety checks. |
| `SSH_MCP_COMMAND_ALLOW` | empty | Command allow patterns. |
| `SSH_MCP_COMMAND_DENY` | empty | Command deny patterns. |
| `SSH_MCP_PATH_ALLOW_PREFIXES` | `/tmp,/var/tmp,/home,/Users` | Remote path prefixes allowed by filesystem policy. |
| `SSH_MCP_PATH_DENY_PREFIXES` | `/etc/sudoers,/etc/shadow,/etc/passwd,/boot,/dev,/proc` | Remote path prefixes denied by filesystem policy. |
| `SSH_MCP_LOCAL_PATH_ALLOW_PREFIXES` | OS temp directory | Local paths allowed for transfer operations. |
| `SSH_MCP_LOCAL_PATH_DENY_PREFIXES` | empty | Local paths denied for transfer operations. |
| `SSH_MCP_TUNNEL_ALLOW_BIND_HOSTS` | `127.0.0.1,localhost,::1` | Local bind hosts allowed for tunnels. |
| `SSH_MCP_TUNNEL_DENY_BIND_HOSTS` | `0.0.0.0,::` | Local bind hosts denied for tunnels. |
| `SSH_MCP_TUNNEL_ALLOW_REMOTE_HOSTS` | empty | Optional remote tunnel target host allowlist. |
| `SSH_MCP_TUNNEL_DENY_REMOTE_HOSTS` | empty | Optional remote tunnel target host denylist. |
| `SSH_MCP_TUNNEL_ALLOW_PORTS` | empty | Optional tunnel port allowlist. |
| `SSH_MCP_TUNNEL_DENY_PORTS` | empty | Optional tunnel port denylist. |
| `SSH_MCP_HTTP_HOST` | `127.0.0.1` | Streamable HTTP bind host. |
| `SSH_MCP_HTTP_PORT` | `3000` | Streamable HTTP bind port. |
| `SSH_MCP_HTTP_ALLOWED_ORIGINS` | `http://127.0.0.1,http://localhost` | Browser origins allowed for HTTP clients. |
| `SSH_MCP_HTTP_BEARER_TOKEN_FILE` | unset | Bearer token file for HTTP transport. Required for non-loopback bearer deployments. |
| `SSH_MCP_ENABLE_LEGACY_SSE` | `false` | Enables legacy SSE compatibility. |
| `SSH_MCP_HTTP_MAX_REQUEST_BODY_BYTES` | `1048576` | Maximum HTTP request body size. |
| `SSH_MCP_HTTP_MAX_SESSIONS` | `20` | Maximum active Streamable HTTP MCP sessions. Expired sessions are cleaned first; if capacity is still full, the oldest idle session is evicted so abandoned clients do not cause persistent 502s. Use `100` for ChatGPT/Cloudflare production deployments. |
| `SSH_MCP_HTTP_SESSION_IDLE_TTL_MS` | `900000` | HTTP MCP session idle timeout in milliseconds. Use `300000` for ChatGPT/Cloudflare production deployments where clients may abandon sessions without DELETE. |
| `SSH_MCP_HTTP_PUBLIC_URL` | unset | Stable public HTTPS MCP URL for protected resource metadata. |
| `SSH_MCP_HTTP_TRUST_PROXY` | `false` | Trust reverse proxy forwarded headers. |
| `SSH_MCP_TOOL_PROFILE` | `full` | Active tool exposure profile. |
| `SSH_MCP_CONNECTOR_PROFILE` | `full` | Alias for `SSH_MCP_TOOL_PROFILE`. |
| `SSH_MCP_CONNECTOR_CREDENTIAL_PROVIDER` | `none` | Credential provider: `none`, `agent`, or `command`. |
| `SSH_MCP_CONNECTOR_CREDENTIAL_COMMAND` | unset | External credential command when provider is `command`. |
| `SSH_MCP_CONNECTOR_CREDENTIAL_COMMAND_ARGS` | empty | Arguments passed to the external credential command. |
| `SSH_MCP_CONNECTOR_CREDENTIAL_COMMAND_TIMEOUT_MS` | `5000` | Credential command timeout in milliseconds. |
| `SSH_MCP_CONNECTOR_DEFAULT_USERNAME` | unset | Default username for connector broker flows. |
| `SSH_MCP_HTTP_AUTH_MODE` | `bearer` | HTTP auth mode: `bearer` or `oauth`. |
| `SSH_MCP_OAUTH_ISSUER` | unset | Expected OAuth issuer. |
| `SSH_MCP_OAUTH_AUDIENCE` | unset | Expected OAuth audience. |
| `SSH_MCP_OAUTH_JWKS_URL` | unset | OAuth JWKS URL. |
| `SSH_MCP_OAUTH_RESOURCE` | unset | OAuth protected resource identifier. |
| `SSH_MCP_OAUTH_REQUIRED_SCOPES` | `ssh-mcp-pro.read` | Required OAuth scopes. |
| `SSH_MCP_OAUTH_ALLOWED_ALGORITHMS` | unset | Optional comma-separated JWT algorithm allowlist, for example `RS256,ES256`. When unset, the built-in OAuth verifier defaults are used. |
| `SSH_MCP_REVIEWER_USERNAME` | unset | Optional dedicated OpenAI reviewer username. Reviewer access remains disabled unless the matching scrypt hash is also configured. |
| `SSH_MCP_REVIEWER_PASSWORD_HASH` | unset | Optional scrypt hash for the dedicated OpenAI reviewer account; never store the plaintext password here. See [OpenAI Reviewer Sandbox](docs/openai-reviewer-sandbox.md). |
| `SSH_MCP_REMOTE_AGENT_MCP_PASSTHROUGH` | unset | When enabled with `1`, `true`, `yes`, or `on`, lets `/mcp` requests bypass the remote control plane and reach the Streamable HTTP MCP handler. Use only for connector routing migrations. |
The parser also accepts non-`SSH_MCP_*` compatibility aliases `PORT`, `KNOWN_HOSTS_PATH`, and `STRICT_HOST_KEY_CHECKING`.
## Tool Profiles
`full` exposes every registered tool, resource, and prompt. Every other profile uses an explicit per-profile allowset. `chatgpt` and `claude` currently expose the same baseline connector tools as `remote-safe`, with empty client-specific extension sets reserved for future additions.
| Profile | Exposed tools | Exposed resources | Exposed prompts |
| --- | --- | --- | --- |
| `full` | All SSH, process, filesystem, transfer, ensure, tunnel, connector, and system tools. | All runtime resources. | All prompts. |
| `remote-safe` | `connector_status`, `ssh_hosts_list`, `ssh_policy_explain`, `ssh_host_inspect`, `ssh_mutation_plan`. | `ssh-mcp-pro://capabilities/support-matrix`. | `inspect-host-capabilities`, `plan-mutation`. |
| `chatgpt` | Baseline remote connector tools plus an empty ChatGPT extension set. | Same remote connector subset as `remote-safe`. | Same remote connector subset as `remote-safe`. |
| `claude` | Baseline remote connector tools plus an empty Claude extension set. | Same remote connector subset as `remote-safe`. | Same remote connector subset as `remote-safe`. |
| `remote-readonly` | Same remote connector subset as `remote-safe`. | Same remote connector subset as `remote-safe`. | Same remote connector subset as `remote-safe`. |
| `remote-broker` | Same remote connector subset as `remote-safe`. | Same remote connector subset as `remote-safe`. | Same remote connector subset as `remote-safe`. |
`SSH_MCP_CHATGPT_EXTRA_TOOLS` and `SSH_MCP_CLAUDE_EXTRA_TOOLS` can extend their profiles for loopback-only development. Non-loopback HTTP startup validates the fully resolved tool set against the canonical remote-safe allowlist and refuses unknown, session, transfer, filesystem mutation, process, sudo, service mutation, or tunnel tools. The startup error lists only the rejected tool names. Adding an extension locally does not make that tool approved for public connector exposure.
## Security Defaults
ssh-mcp-pro starts with strict SSH host-key verification, denies root login, denies raw sudo, blocks destructive commands and filesystem operations unless policy allows them, and refuses non-loopback HTTP startup unless authentication, origins, public HTTPS URL, strict host-key verification, a remote-safe tool profile, and host allowlists are configured. See [SECURITY.md](SECURITY.md) for vulnerability reporting and [SECURITY_DECISIONS.md](SECURITY_DECISIONS.md) for the design rationale behind these defaults.
## More Documentation
- [INSTALL.md](INSTALL.md) covers full client setup and troubleshooting.
- [API reference](https://oaslananka.github.io/ssh-mcp-pro/) is generated from the published TypeScript entry points.
- [CHANGELOG.md](CHANGELOG.md) records release history in Keep a Changelog format.
- [AGENTS.md](AGENTS.md) describes agent-facing operational guidance.
- [examples/README.md](examples/README.md) contains workflow examples.
- [ARCHITECTURE.md](ARCHITECTURE.md) explains the major subsystems and ADRs.
- [REGISTRY_SUBMISSION.md](REGISTRY_SUBMISSION.md) tracks MCP Registry submission readiness.
- [docs/tutorials/getting-started.md](docs/tutorials/getting-started.md),
[docs/how-to/](docs/how-to/README.md), [docs/reference/](docs/reference/README.md),
and [docs/explanation/](docs/explanation/architecture.md) organize the docs above by
task (tutorial, how-to, reference, explanation).
- [docs/troubleshooting.md](docs/troubleshooting.md) covers common failure modes.
## Project Health & Governance
- [GOVERNANCE.md](GOVERNANCE.md) describes how decisions are made today.
- [MAINTAINERS.md](MAINTAINERS.md) lists current maintainers.
- [ROADMAP.md](ROADMAP.md) tracks known process gaps and what's planned to close them.
- [CODE_OF_CONDUCT.md](CODE_OF_CONDUCT.md) sets community expectations.
- [docs/repo-maturity-report.md](docs/repo-maturity-report.md) is an evidence-based
audit of this repository's open-source and OpenSSF maturity, including what's not
yet in place.
## Contributing
See [CONTRIBUTING.md](CONTRIBUTING.md) for setup, quality gates, commit rules,
and pull request expectations. Participation is governed by the
[Code of Conduct](CODE_OF_CONDUCT.md).
## License
ssh-mcp-pro project-authored source is available under [GNU AGPL v3.0](LICENSE).
## Agent plugin and runtime configuration
This repository owns the product-level agent plugin, MCP runtime configuration, and product-specific skills for `ssh-mcp-pro`. The central [`agent-tools`](https://github.com/oaslananka/agent-tools) repository should catalog this plugin, but the manifest and workflow instructions live here so they stay synchronized with the actual MCP server package.
| File | Purpose |
| --- | --- |
| [`.claude-plugin/plugin.json`](.claude-plugin/plugin.json) | Claude Code-valid product plugin manifest. |
| [`.mcp.json`](.mcp.json) | Claude Code project-local MCP server configuration. |
| [`.codex/config.example.toml`](.codex/config.example.toml) | Codex CLI MCP configuration example. |
| [`.vscode/mcp.example.json`](.vscode/mcp.example.json) | VS Code / GitHub Copilot workspace MCP configuration example. |
| [`opencode.example.jsonc`](opencode.example.jsonc) | OpenCode project MCP configuration example. |
| `.opencode/skills/` | OpenCode-native mirrored skill definitions. |
| [`docs/agent-runtime-config.md`](docs/agent-runtime-config.md) | Agent runtime setup and validation notes. |
Validate plugin packaging locally:
```bash
claude plugin validate .
```
TDQS
Scored across 46 tools
Several tool pairs have blurred boundaries: ssh_list_configured_hosts vs ssh_hosts_list both enumerate hosts, log_tail vs service_logs both read journal entries for a service, and ensure_service overlaps with service_restart/service_stop/service_status. proc_sudo vs proc_exec and ssh_host_inspect vs ssh_resolve_host also require the agent to reason about subtle policy differences rather than purpose.
The set follows a clear domain-prefix convention (fs_*, ssh_*, service_*, proc_*, tunnel_*, file_*), which makes names scannable and mostly predictable. Minor deviations exist: get_metrics uses verb_noun while peers use noun_verb, os_detect is noun_verb amid prefix-first names, and ssh_list_sessions vs ssh_hosts_list invert the ordering within the same family.
46 tools is heavy for what is essentially remote host administration, and the surface shows signs of accretion (multiple logging tools, duplicate host-listing tools, both proc_exec and proc_exec_stream, both connector_status and get_metrics). A single SSH server would be more legible with roughly half this count by merging overlapping families.
Coverage is broad for the domain: session lifecycle, filesystem CRUD and stats, package/service management, process control, tunnels, transfers, metrics, and a policy/planning layer. Gaps are minor — no user/group or cron/scheduled-task management, and no explicit package listing, but agents can reach those via proc_exec.