fix: remove sandbox language from tool descriptions
The terminal and execute_code tool schemas unconditionally mentioned 'cloud sandboxes' in their descriptions sent to the model. This caused agents running on local backends to believe they were in a sandboxed environment, refusing networking tasks and other operations. Worse, agents sometimes saved this false belief to persistent memory, making it persist across sessions. Reported by multiple users (XLion, 林泽).
This commit is contained in:
@@ -531,7 +531,6 @@ Working directory: Use 'workdir' for per-command cwd.
|
||||
PTY mode: Set pty=true for interactive CLI tools (Codex, Claude Code, Python REPL).
|
||||
|
||||
Do NOT use vim/nano/interactive tools without pty=true — they hang without a pseudo-terminal. Pipe git output to cat if it might page.
|
||||
Important: cloud sandboxes may be cleaned up, idled out, or recreated between turns. Persistent filesystem means files can resume later; it does NOT guarantee a continuously running machine or surviving background processes. Use terminal sandboxes for task work, not durable hosting.
|
||||
"""
|
||||
|
||||
# Global state for environment lifecycle management
|
||||
|
||||
Reference in New Issue
Block a user