aiterm-mcp
Persistent terminal MCP with one harness-based launcher for Claude Code, Codex CLI, Grok CLI, and Cursor Agent CLI, plus durable PTYs for SSH, containers, and REPLs.
npm · fingerprint 7a2e0963557602ca05b1f20d · repository · RSS
16 tools
1 recorded versions
13h tracked
tracking started
What changed Nothing, since tracking began. That is the good outcome, and it is what most servers look like.
Current tools
Show all 16 tool fingerprints
agent_configure
3fb09b27a363eb92
agent_launch
e4dfaa40b00ccd7f
agent_steer
83deba9e37935bf9
claude_agent
c1710095aca8a008
claude_approval
c6070e36454a781f
claude_turn
dc12e6a695a2db74
codex_agent
92a428e96e573ed3
composer_agent
f2160c7d9c023119
diagnostics
aaa6d8ad8e62ff60
grok_agent
62815854a688609c
pty_close
b2a2fc5c6706d9e0
pty_list
dedb1b517991c4c4
pty_open
750d683c72ba0573
pty_read
fb3d4dda88b61314
pty_send
c5817199048322fc
Watch this server yourself
If you run this server, put the proxy in front of it. It pins these exact
fingerprints on first connect and stops the session if they move.
npx --yes mcp-pin@0.1.0 -- <your aiterm-mcp command>
Or subscribe to this page's RSS feed
to be told when it changes.
Badge
The badge states one fact about time and nothing else. It never claims a
server is safe.
[](https://mcp-pin.gautamkhosla.com/servers/44524933e873a07e.html)
mcp-pin keeps a public, append-only record of MCP tool definitions. Every entry is hash linked and every head is signed, so you can download the log and check it yourself with npx --yes mcp-pin@0.1.0 verify-log. You do not have to trust whoever runs this.
Crawling is one tools/list per server per day. No tool is ever called. To opt out, add your server to OPTOUT.txt or open an issue. Honoured on the next crawl, no justification needed.
Run by Gautam Khosla as an independent open-source project. Not affiliated with
Anthropic, the Model Context Protocol project, or any server listed here.
About this project, and how to contact me .
MIT licensed. Source on GitHub .