diff --git a/docs/src/datman/getting_started.md b/docs/src/datman/getting_started.md
index 04807b5..1e358cb 100644
--- a/docs/src/datman/getting_started.md
+++ b/docs/src/datman/getting_started.md
@@ -12,9 +12,17 @@ Run `datman init` in an empty directory.
 `datman.toml` will be created.
 
 
-## Adding a Source
+## Adding a Directory Source
 
-TODO
+Add a section like the following to your `datman.toml`:
+
+```toml
+[source.mysourcename]
+directory = "/path/to/source"
+hostname = "myhostname"
+```
+
+The `hostname` field is currently unused but may be used in the future. It should be set to the hostname of the machine where the data resides.
 
 
 ## Begin Labelling
@@ -38,4 +46,33 @@ Numbers assign a literal label to the file in question.
 
 ## Interactively Browse Tree
 
-Run `datman ibrowse` to interactively browse the tree.
+Run `datman ibrowse [source]` to interactively browse the tree.
+
+
+## Adding a Destination
+
+Change to a different directory and run `yama init` to create a Yama pile.
+
+Then edit your `datman.toml` and add:
+
+```toml
+[piles.mypilename]
+path = "path/to/my/pile"
+included_labels = ["precious"]
+```
+
+Include whichever labels you would like to include in the backup to that destination.
+
+
+## Running a Backup
+
+Run `datman backup-all [destination name]` or `datman backup-one [source name] [destination name]`.
+
+
+## Extracting a Backup
+
+Run `datman extract --accept-partial [pile name] [destination directory of extraction, will be created]`.
+
+If not running as root, add `--skip-metadata` to prevent failing on `chown`.
+
+To extract the first backup of each source before a certain date/datetime, use `--before [date/datetime]`. Similar, intuitive behaviour exists for `--after [date/datetime]`.
diff --git a/docs/src/yama/getting_started.md b/docs/src/yama/getting_started.md
index 0d1e551..7a0ab09 100644
--- a/docs/src/yama/getting_started.md
+++ b/docs/src/yama/getting_started.md
@@ -50,3 +50,14 @@ Then use `yama retrieve-adv (pointer ID) dest-path` to retrieve files from the p
 
 
 TODO this guide still needs work. Once yama is done.
+
+
+## Check
+
+Use `yama check --shallow` to do a shallow (fast) check.
+
+`yama check --deep` does a deep check, by reading in every single chunk.
+
+Adding `--dry-run-gc` will report the number of unused chunks. There is also an option to sweep these away, which can be found in `--help` but is not yet documented here.
+
+Executing garbage collection is not yet recommended.