Skip to content

ccproxy.cli.commands.permission

ccproxy.cli.commands.permission

MCP permission prompt tool for Claude Code SDK.

permission_tool 

permission_tool(tool_name, tool_input, config=None)

MCP permission prompt tool for Claude Code SDK.

This tool is used by the Claude Code SDK to check permissions for tool calls. It returns a JSON response indicating whether the tool call should be allowed or denied.

Response format: - Allow: {"behavior": "allow", "updatedInput": {...}} - Deny: {"behavior": "deny", "message": "reason"}

Examples:

ccproxy-perm "bash" '{"command": "ls -la"}' ccproxy-perm "edit_file" '{"path": "/etc/passwd", "content": "..."}'

Source code in ccproxy/cli/commands/permission.py 
 27
 28
 29
 30
 31
 32
 33
 34
 35
 36
 37
 38
 39
 40
 41
 42
 43
 44
 45
 46
 47
 48
 49
 50
 51
 52
 53
 54
 55
 56
 57
 58
 59
 60
 61
 62
 63
 64
 65
 66
 67
 68
 69
 70
 71
 72
 73
 74
 75
 76
 77
 78
 79
 80
 81
 82
 83
 84
 85
 86
 87
 88
 89
 90
 91
 92
 93
 94
 95
 96
 97
 98
 99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
@app.command()
def permission_tool(
    tool_name: Annotated[
        str, typer.Argument(help="Name of the tool to check permissions for")
    ],
    tool_input: Annotated[str, typer.Argument(help="JSON string of the tool input")],
    config: Annotated[
        Path | None,
        typer.Option(
            "--config",
            "-c",
            help="Path to configuration file (TOML, JSON, or YAML)",
            exists=True,
            file_okay=True,
            dir_okay=False,
            readable=True,
        ),
    ] = None,
) -> None:
    """
    MCP permission prompt tool for Claude Code SDK.

    This tool is used by the Claude Code SDK to check permissions for tool calls.
    It returns a JSON response indicating whether the tool call should be allowed or denied.

    Response format:
    - Allow: {"behavior": "allow", "updatedInput": {...}}
    - Deny: {"behavior": "deny", "message": "reason"}

    Examples:
        ccproxy-perm "bash" '{"command": "ls -la"}'
        ccproxy-perm "edit_file" '{"path": "/etc/passwd", "content": "..."}'
    """

    try:
        # Parse the tool input JSON
        try:
            input_data = json.loads(tool_input)
        except json.JSONDecodeError as e:
            response = PermissionToolDenyResponse(message=f"Invalid JSON input: {e}")
            print(response.model_dump_json(by_alias=True))
            raise typer.Exit(1) from e

        # Load settings to get permission configuration
        settings = config_manager.load_settings(config_path=config)

        # Basic permission checking logic
        # This can be extended with more sophisticated rules

        # Check for potentially dangerous commands
        dangerous_patterns = [
            "rm -rf",
            "sudo",
            "passwd",
            "chmod 777",
            "/etc/passwd",
            "/etc/shadow",
            "format",
            "mkfs",
        ]

        # Convert input to string for pattern matching
        input_str = json.dumps(input_data).lower()

        # Check for dangerous patterns
        for pattern in dangerous_patterns:
            if pattern in input_str:
                response = PermissionToolDenyResponse(
                    message=f"Tool call contains potentially dangerous pattern: {pattern}"
                )
                print(response.model_dump_json(by_alias=True))
                return

        # Check for specific tool restrictions
        restricted_tools = {"exec", "system", "shell", "subprocess"}

        if tool_name.lower() in restricted_tools:
            response = PermissionToolDenyResponse(
                message=f"Tool {tool_name} is restricted for security reasons"
            )
            print(response.model_dump_json(by_alias=True))
            return

        # Allow the tool call with original input
        allow_response = PermissionToolAllowResponse(updated_input=input_data)
        print(allow_response.model_dump_json(by_alias=True))

    except Exception as e:
        error_response = PermissionToolDenyResponse(
            message=f"Error processing permission request: {e}"
        )
        print(error_response.model_dump_json(by_alias=True))
        raise typer.Exit(1) from e

main 

main()

Entry point for ccproxy-perm command.

Source code in ccproxy/cli/commands/permission.py 
122
123
124
def main() -> None:
    """Entry point for ccproxy-perm command."""
    app()