Docs: Improve agent propagation and add running manually page

docs/content/reference/agent_propagation.md

@@ -9,7 +9,7 @@ tags: ["agent", "propagation", "reference"]
 
 ## How does the Infection Monkey Agent propagate to a new machine?
 
-Once an Agent exploits a vulnerable system, it propagates to the machine by copying the appropriate
-Agent binary to it.
+Agent mainly propagates using remote code execution vulnerabilities. Once the agent is able to
+run commands on the victim it executes commands that are similar to the ones described in [manual run page.](../../usage/running-manually/)
 
 On Windows, it is copied to `C:\Windows\temp\monkey64.exe`. On Linux, it is copied to `/tmp/monkey`.

docs/content/usage/integrations/aws-run-on-ec2-machine.md

@@ -1,5 +1,5 @@
 ---
-title: "Running the monkey on AWS EC2 instances"
+title: "Running the agent on AWS EC2 instances"
 date: 2020-06-28T10:44:05+03:00
 draft: false
 description: "Use AWS SSM to execute Infection Monkey on your AWS instances."

docs/content/usage/running-manually.md

@@ -0,0 +1,32 @@
+---
+title: "Running Manually"
+date: 2022-06-09T14:47:40+03:00
+draft: false
+weight: 2
+pre: "<i class='fas fa-terminal'></i> "
+tags: ["usage"]
+---
+
+
+## Generating manual run command
+
+Manual run command can be generated through the Island Server UI, by going to "Run Monkey" -> "Manual" page.
+
+### Downloading the agent manually
+
+As evident by the generated commands, agent binaries can be downloaded
+by sending a get request to `https://[IP]:5000/api/agent/download/[OS]`, where
+`[IP]` stands for the IP address of the Island server and `[OS]` is either `windows` or `linux`.
+
+### Running the agent binary
+
+Agent binary can be started with `m0nk3y` or `dr0pp3r` flags.
+
+`m0nk3y` flag is the standard way
+to run the agent.
+
+`dr0pp3r` will move the agent binary to a location provided with an `-l` flag.
+Then it will start that binary with a `m0nk3y` flag on a new process. Finally, it will stop
+the current process and shut down. This flag is useful if you want to start the agent on a separate
+process. This flag is used by the agent to execute other agents on exploited machines.
+This way, the parent agent doesn't have to wait until the child finishes to continue the execution.