Updated docs

README.md

@@ -1,12 +1,44 @@
 # wptagent
-Cross-platform WebPageTest agent (currently supports Chrome only on Windows and Linux)
+Cross-platform WebPageTest agent
+
+## Supported Platforms/Browsers
+Chrome is the only browser that currently supports capturing response bodies and manipulating requests (changing headers, blocking requests, etc).  All browsers should support basic page loading, scripts and video capture on all platforms.  Traffic-shaping is supported on all platforms as well.
+
+### Linux (with display or headless with Xvfb)
+* Chrome (Stable and Unstable)
+* Firefox (Stable and Nightly)
+
+### Windows
+* Chrome (Stable and Canary)
+* Firefox (Stable and Nightly)
+* Microsoft Edge (does not support complex scripts or waiting for page activity yet, just top-level page navigations)
+
+### OSX
+* Chrome (Stable and Canary)
+* Firefox (Stable and Nightly)
+
+### Android (requires a tethered host - Raspberry Pi's preferred)
+* Chrome (Stable, Beta and Unstable)
+* Several browsers run as "black box" tests (single page load, visual metrics only):
+    * Chrome (Stable, Beta, Dev and Canary)
+    * Opera
+    * Opera Mini
+    * UC Browser
+    * UC Mini
+    * Samsung Browser
+    * Firefox (Stable and Beta)
 
 ## Known Issues
 * Not all features have been implemented yet (see list below)
+* Microsoft Edge does not (yet) support waiting beyond a page's onload event for network activity
 
 ## Installation
 * Install docs are [here](docs/install.md)
 
+## Run with docker
+Check out [the docker instructions](docs/docker.md) for information on how to
+run the agent in a docker container.
+
 ## Command-line options
 ### Basic agent config
 * **-v** : Increase verbosity (specify multiple times for more). -vvvv for full debug output.
@@ -110,6 +142,3 @@ Cross-platform WebPageTest agent (currently supports Chrome only on Windows and
     * overrideHost (depends on support being added to dev tools)
     * if/else/endif
 
-## Run with docker
-Check out [the docker instructions](docs/docker.md) for information on how to
-run the agent in a docker container.

docs/install.md

@@ -1,5 +1,5 @@
 # Requirements
-wptagent currently supports Windows and Linux hosts (possibly OSX but not tested).  If traffic-shaping is enabled the agent will need to be able to elevate to admin/root. It is recommended that the agent itself not run as admin/root but that it can elevate without prompting which means disabling UAC on windows or adding the user account to the sudoers file on Linux.
+wptagent currently supports Windows and Linux hosts (possibly OSX but not tested).  If traffic-shaping is enabled the agent will need to be able to elevate to admin/root. It is recommended that the agent itself not run as admin/root but that it can elevate without prompting which means disabling UAC on windows or adding the user account to the sudoers file on Linux and OSX (NOPASSWD in visudo).
 
 ## Software Requirements
 * Python 2.7 available on the path (python2.7 and  python-pip packages on Ubuntu/Debian) with the following modules installed (all available through pip):
@@ -51,7 +51,7 @@ wptagent currently supports Windows and Linux hosts (possibly OSX but not tested
     * ```sudo npm install -g lighthouse```
 
 ## Configuration
-The default browser locations for Chrome and Firefox (Stable, Beta and Canary/Nightly) will automatically be detected.  If you need to support different locations or provide different browsers you can rename browsers.ini.sample to browsers.ini and define the browser locations.
+The default browser locations for Chrome, Firefox (Stable, Beta and Canary/Nightly) and Microsoft Edge will automatically be detected.  If you need to support different locations or provide different browsers you can rename browsers.ini.sample to browsers.ini and define the browser locations.
 
 On Linux, make sure to point to the actual chrome binary and not the symlink.  Usually something like:
 ```