[KongchangAI]
ProductAI SSH Pro

AI-SSH-Pro

A local multi-session SSH desktop client with a built-in AI operations assistant governed by a controllable execution policy. It supports OpenAI-compatible models and is open-sourced under the MIT license.

Timeline (last 90 days)

Oct 20

AI-SSH-Pro is a local, multi-session SSH desktop client, open-sourced under the MIT license, with Chinese and English interfaces.

Unverified50%
Oct 20

AI-SSH-Pro supports importing configuration from ~/.ssh/config, multi-level ProxyJump bastion hosts, port forwarding and host key verification.

Unverified50%
Oct 20

AI-SSH-Pro's AI execution policies come in three tiers: Suggest only, Confirm before run (default), and Auto run.

Unverified50%
Oct 20

Under the Confirm before run policy, read-only commands execute directly, change-type commands require confirmation, and high-risk commands require secondary confirmation.

Unverified50%
Oct 20

Under the Auto run policy, low- and medium-risk commands execute automatically while high-risk commands are still queried, and automatic execution of high-risk commands can additionally be enabled (off by default).

Unverified50%
Oct 20

AI-SSH-Pro supports any OpenAI-compatible model provider, such as OpenAI, DeepSeek and Qwen, with the API Key stored on the local machine.

Unverified50%
Oct 20

AI-SSH-Pro's write restriction for read-only hosts is a hard boundary: the main process re-verifies the host's protection status before the AI writes to the terminal, and on a read-only host no write is possible even if the user clicks approve.

Unverified50%
Oct 20

In AI-SSH-Pro, production hosts are not forcibly downgraded; the policy chosen by the user takes effect as-is, and the banner only serves as a reminder; only read-only hosts actually prevent AI writes.

Unverified50%
Oct 20

Each terminal tab in AI-SSH-Pro corresponds to an independent session, with two tabs on the same host shown as #1 and #2; the AI executes no commands when no terminal is bound.

Unverified50%
Oct 20

AI-SSH-Pro provides dual-pane SFTP, resource charts and multi-host monitoring, alert rules and an alert workbench, Docker container and Compose management, Playbooks, batch execution, session recording and playback, and audit logs.

Unverified50%

2 more timeline events

All Facts (12)

Unverified

AI-SSH-Pro is a local, multi-session SSH desktop client, open-sourced under the MIT license, with Chinese and English interfaces.

50%
Unverified

AI-SSH-Pro supports importing configuration from ~/.ssh/config, multi-level ProxyJump bastion hosts, port forwarding and host key verification.

50%
Unverified

AI-SSH-Pro's AI execution policies come in three tiers: Suggest only, Confirm before run (default), and Auto run.

50%
Unverified

Under the Confirm before run policy, read-only commands execute directly, change-type commands require confirmation, and high-risk commands require secondary confirmation.

50%
Unverified

Under the Auto run policy, low- and medium-risk commands execute automatically while high-risk commands are still queried, and automatic execution of high-risk commands can additionally be enabled (off by default).

50%
Unverified

AI-SSH-Pro supports any OpenAI-compatible model provider, such as OpenAI, DeepSeek and Qwen, with the API Key stored on the local machine.

50%
Unverified

AI-SSH-Pro's write restriction for read-only hosts is a hard boundary: the main process re-verifies the host's protection status before the AI writes to the terminal, and on a read-only host no write is possible even if the user clicks approve.

50%
Unverified

In AI-SSH-Pro, production hosts are not forcibly downgraded; the policy chosen by the user takes effect as-is, and the banner only serves as a reminder; only read-only hosts actually prevent AI writes.

50%
Unverified

Each terminal tab in AI-SSH-Pro corresponds to an independent session, with two tabs on the same host shown as #1 and #2; the AI executes no commands when no terminal is bound.

50%
Unverified

AI-SSH-Pro provides dual-pane SFTP, resource charts and multi-host monitoring, alert rules and an alert workbench, Docker container and Compose management, Playbooks, batch execution, session recording and playback, and audit logs.

50%
Unverified

AI-SSH-Pro's tab-level policy overrides exist only in memory and are reset on close or disconnect, and they still respect read-only restrictions.

50%
Unverified

AI-SSH-Pro has a built-in demo host with simulated data, letting users try it without touching a real server.

50%

Source Articles