microsoft · johnstairs · Jun 2, 2026 · Jun 2, 2026
diff --git a/cli/internal/install/cloudinstall/cloud-config-pretty.tpl b/cli/internal/install/cloudinstall/cloud-config-pretty.tpl
@@ -61,7 +61,7 @@ cloud:
     # name:
   {{- end }}
 
-  # Optionaly use a custom DNS zone
+  # Optionally use a custom DNS zone
   {{ if .DnsZone -}}
   dnsZone:
     resourceGroup: {{ .DnsZone.ResourceGroup }}

diff --git a/deploy/config/microsoft/cloudconfig-private-link.yml b/deploy/config/microsoft/cloudconfig-private-link.yml
@@ -38,7 +38,7 @@ cloud:
     resourceGroup: eminence
     name: eminence
 
-  # Optionaly use a custom DNS zone
+  # Optionally use a custom DNS zone
   # dnsZone:
   # resourceGroup:
   # name:

diff --git a/deploy/config/microsoft/cloudconfig.yml b/deploy/config/microsoft/cloudconfig.yml
@@ -38,7 +38,7 @@ cloud:
     resourceGroup: eminence
     name: eminence
 
-  # Optionaly use a custom DNS zone
+  # Optionally use a custom DNS zone
   dnsZone:
     resourceGroup: eminence
     name: eminence.ms

diff --git a/docs/guides/buffers.md b/docs/guides/buffers.md
@@ -26,13 +26,13 @@ buffer reading, buffer writing, and creating runs.
 
 In a cloud installation, you can use multiple storage accounts for buffer
 storage. If those accounts are in different cloud locations (regions), you can
-specify the location that that you would like the buffer to be crated in with
+specify the location that you would like the buffer to be created in with
 the `--location` parameter. The location is the lowercase name of the cloud
 location with spaces removed, such as `eastus` or `westus2`. If multiple storage
 accounts are in the same location, accounts are selected in a round-robin
 fashion.
 
-The available regions for a tyger installation can fetched with:
+The available regions for a Tyger installation can be fetched with:
 
 ```bash
 tyger buffer storage-account list
@@ -63,7 +63,7 @@ The `--flush-interval` parameter specifies the maximum time interval that data
 You can specify the block size with `--block-size`, for example,
 `--block-size 16M`. The default block size is 4MB.
 
-Instead of standard in, you can use `-i|--input` to read from a file or named
+Instead of standard input, you can use `-i|--input` to read from a file or named
 pipe.
 
 ## Reading from buffers
@@ -270,7 +270,7 @@ By default, active buffers do not have an expiration date and will never be auto
 This behavior is configured during Tyger installation via the `buffers.activeLifetime` field.
 If an active buffer expires, it will be automatically soft-deleted by Tyger.
 
-To set the TTL for a buffer manually, use the format `DD.HH:MM:SS`.
+To set the TTL for a buffer manually, use the format `D.HH:MM:SS`.
 
 ```bash
 tyger buffer set --ttl 2.12:00 $buffer_id --soft-deleted
@@ -360,7 +360,7 @@ If multiple storage accounts are configured for the source Tyger installation
 from a single storage account. The available storage accounts can be listed with
 the command `tyger buffer storage-account list`.
 
-To only export a subset of buffer, you can filter the buffers to be exported by
+To only export a subset of buffers, you can filter the buffers to be exported by
 tags.
 
 This command starts a special [run](./runs). Logs are displayed inline, but can also be
@@ -374,11 +374,11 @@ into the destination Tyger instance's database with the command:
 tyger buffer import [--storage-account STORAGE_ACCOUNT_NAME]
 ```
 
-This starts a run that scans though the instance's storage account and imports
+This starts a run that scans through the instance's storage account and imports
 new buffers. Note that existing buffers are not touched and their tags will not
 be updated.
 
 If multiple storage accounts are configured for the Tyger installation
 `--storage-account` must be provided. The import command will only import from a
-from a single storage account. The available storage accounts can be listed with
+single storage account. The available storage accounts can be listed with
 the command `tyger buffer storage-account list`.
diff --git a/docs/guides/codespecs.md b/docs/guides/codespecs.md
@@ -1,7 +1,7 @@
 # Working with codespecs
 
 Codespecs in Tyger define the code executed during a run. Named codespecs can be
-used by for multiple runs. While can also declare codespecs inline, this guide
+used for multiple runs. While you can also declare codespecs inline, this guide
 focuses on working with named codespecs.
 
 ## Creating a codespec
@@ -35,10 +35,11 @@ buffers:
     - output
 image: quay.io/linuxserver.io/ffmpeg
 args:
+  - -i
   - $(INPUT_PIPE)
   - -vf
   - negate
-  - -f-
+  - -f
   - nut
   - -y
   - $(OUTPUT_PIPE)
@@ -58,7 +59,7 @@ tyger codespec create -f negating.yml --env MY_ENV=MY_VALUE
 
 ## Using buffers
 
-The commands above specify two buffers one for input, one for output, named
+The commands above specify two buffers, one for input and one for output, named
 `input` and `output` respectively. For each buffer, tyger creates a named pipe
 for reading or writing buffer contents, depending on the declared
 directionality. The path to the named pipe is given in an environment variable
@@ -100,7 +101,7 @@ image: quay.io/linuxserver.io/ffmpeg
 # Entrypoint array. Not executed within a shell.
 # The container image's ENTRYPOINT is used if this is not provided.
 # Variable references $(VAR_NAME) are expanded using the container's
-# environment.The $(VAR_NAME) syntax can be escaped with a double $$, ie: $$(VAR_NAME).
+# environment. The $(VAR_NAME) syntax can be escaped with a double $$, ie: $$(VAR_NAME).
 command:
   - ffmpeg
 
@@ -110,10 +111,11 @@ command:
 # the reference in the input string will be unchanged. The $(VAR_NAME)
 # syntax can be escaped with a double $$, ie: $$(VAR_NAME).
 args:
+  - -i
   - $(INPUT_PIPE)
   - -vf
   - negate
-  - -f-
+  - -f
   - nut
   - -y
   - $(OUTPUT_PIPE)
@@ -131,7 +133,7 @@ env:
   MY_VAR: myValue
 
 # An optional workload identity to run as, allowing access to cloud resources.
-# This must be one of the identites given in the cloud configuration file under
+# This must be one of the identities given in the cloud configuration file under
 # cloud.compute.identities.
 identity: my-identity
 
@@ -208,14 +210,14 @@ specified, in which case they are treated as the `command` value.
 Retrieve a specific codespec version with:
 
 ```bash
-tyger codespec show NAME [-v|--version VERSION]
+tyger codespec show NAME [--version VERSION]
 ```
 
 Without `--version`, the latest version is returned.
 
 ## Listing codespecs
 
-List **latest version** of codespecs with:
+List the **latest version** of codespecs with:
 
 ```bash
 tyger codespec list [--prefix STRING] [--limit COUNT]

diff --git a/docs/guides/distributed-runs.md b/docs/guides/distributed-runs.md
@@ -9,7 +9,7 @@ environment.
 
 All runs in Tyger use a "job" codespec for primary execution. Distributed runs
 additionally employ workers for distributing workloads. These workers, defined
-by a separate codespec, typically listen on network ports so that the the job
+by a separate codespec, typically listen on network ports so that the job
 can communicate with them over the cluster's local network.
 
 ## Creating a worker codespec
@@ -82,5 +82,5 @@ JSON array of `hostname:port` strings.
 ## Example
 
 [Gadgetron
-examples](../reference/gadgetron/gadgetron.md#distributed-reconstruction) has an
-example that uses a disributed run.
+examples](../reference/gadgetron/gadgetron.md#distributed-reconstruction) include an
+example that uses a distributed run.
diff --git a/docs/guides/ephemeral-buffers.md b/docs/guides/ephemeral-buffers.md
@@ -9,7 +9,7 @@ Tyger.
 
 While buffers are very powerful, there are cases when you don't want data to be
 persisted and/or you want low-latency communication between a data source and a
-run's container. For these scenario's you can use **ephemeral buffers**.
+run's container. For these scenarios you can use **ephemeral buffers**.
 
 When using ephemeral buffers, instead of data being sent to the data plane
 service (Azure Storage or the local data plane service), a sidecar container to
@@ -21,7 +21,7 @@ However, there are a few important differences between the two:
 - Ephemeral buffers can only be read once, and therefore are not WORM.
 - Ephemeral buffers are not created with `tyger buffer create`. Rather, they are
   created with, and are owned by, a run.
-- You can only read or write to an ephemeral buffer during while its associated
+- You can only read or write to an ephemeral buffer while its associated
   run is active.
 
 ## Using ephemeral buffers

diff --git a/docs/guides/login.md b/docs/guides/login.md
@@ -18,7 +18,7 @@ tyger login SERVER_URL [--use-device-code] [--proxy PROXY]
 ```
 
 This launches a browser tab for interactive login. If this isn't
-possible, use `--use-device-code` to receive a device code and and manually open a
+possible, use `--use-device-code` to receive a device code and manually open a
 provided URL for authentication.
 
 The `--proxy` option allows specifying an HTTP proxy for all HTTP requests,
@@ -37,13 +37,13 @@ system's certificate store.
 tyger login
     SERVER_URL
     --service-principal APPID
-    --certificate CERTPATH | --cert-thumbprint THUMBPRINT
+    --cert-file CERTPATH | --cert-thumbprint THUMBPRINT
     [--proxy PROXY]
 ```
 
 ## Log in using a managed identity
 
-If you are running on an Azure VM, you can login to tyger using a managed identity with:
+If you are running on an Azure VM, you can log in to Tyger using a managed identity with:
 
 ```bash
 tyger login SERVER_URL --identity [--identity-client-id MI_ID] [--federated-identity TARGET_CLIENT_ID]
@@ -59,7 +59,7 @@ specify the client ID of the target identity with `--federated-identity`.
 ## Log in from GitHub Actions
 
 Similar to Azure managed identities, you can use [federated
-credentials](https://learn.microsoft.com/en-us/entra/workload-id/workload-identity-federation-create-trust?pivots=identity-wif-apps-methods-azp#github-actions∑)
+credentials](https://learn.microsoft.com/en-us/entra/workload-id/workload-identity-federation-create-trust?pivots=identity-wif-apps-methods-azp#github-actions)
 to log in from a GitHub Actions runner:
 
 ```bash
@@ -94,7 +94,7 @@ This requires that:
 
 By default, `tyger` uses the operating system's trusted root CA certificates for
 TLS certificate validation. You can override this with the
-`--tls-ca-certificates` flag. The following values are supported:
+`--tls-certificates` flag. The following values are supported:
 
 - `os` to use the operating system's root CA certificates (the default)
 - `embedded` to use the CA certificates embedded in the `tyger` binary. These

diff --git a/docs/guides/runs.md b/docs/guides/runs.md
@@ -16,13 +16,13 @@ When Tyger is running in a local Docker environment, container images used by
 runs need to be pulled in advance using `docker pull`. This is a security
 measure to prevent using Tyger to introduce untrusted container images onto the
 system. The Tyger CLI will pull the image for you if you pass in `--pull` to
-to `tyger run create` or `tyger run exec`.
+`tyger run create` or `tyger run exec`.
 
 :::
 
 ## Creating runs with `exec`
 
-`tyger run exec` is a the easiest way to create and execute a run. It allows
+`tyger run exec` is the easiest way to create and execute a run. It allows
 up to one buffer's contents to be provided through standard input and up to one
 buffer's output to be written to standard output.
 
@@ -56,7 +56,7 @@ echo "Paul" | tyger run exec --codespec hello > result.txt
 ```
 
 This will write status information to standard error and "Hello Paul" to
-results.txt.
+result.txt.
 
 You can also provide the run specification as a file:
 
@@ -113,7 +113,7 @@ tyger buffer read $output_id > result.txt
 ```
 
 Notice how we pass in buffers as arguments to the codespec's buffer parameters.
-Missing buffers arguments are automatically created and their IDs can be retrieved using
+Missing buffer arguments are automatically created and their IDs can be retrieved using
 `tyger run show`.
 
 ## `exec` and `create` options
@@ -168,7 +168,7 @@ cluster: mycluster
 
 # The run is given this amount of time to complete,
 # starting from when the run was created, not when it
-# when it started executing.
+# started executing.
 timeoutSeconds: 43200
 ```
 
@@ -217,7 +217,7 @@ To print the entire resource, specify `--full-resource`.
 ## Tagging runs
 
 Runs can be tagged with key-value metadata pairs just like
-[buffers](buffers.html#tagging-buffers).You can assign tags to a
+[buffers](buffers.html#tagging-buffers). You can assign tags to a
 run when creating it like this:
 
 ```bash
@@ -261,11 +261,11 @@ Runs are listed in descending order of creation time. If `--limit` is not
 specified, a maximum of 1000 runs are shown with a warning if the output had to
 be truncated.
 
-Use the `--since` to only include runs that were created after the given time.
+Use `--since` to only include runs that were created after the given time.
 
-Use the `--tag` parameter to restrict the results can contain **all** of the given tags.
+Use the `--tag` parameter to restrict the results to those that contain **all** of the given tags.
 
-Use the `--status` command to restrict the results that have **any** of the given statues.
+Use the `--status` parameter to restrict the results to those that have **any** of the given statuses.
 
 ::: info Tip
 
@@ -275,7 +275,7 @@ Use `tyger run list --limit 1` to fetch the most recent run.
 
 ## Displaying run counts
 
-You can fetch a summary of the counts of runs group by status using:
+You can fetch a summary of the counts of runs grouped by status using:
 
 ```bash
 tyger run counts [--since DATE/TIME] [--tag key=value [...]]
@@ -291,7 +291,7 @@ You can cancel a job with:
 tyger run cancel ID
 ```
 
-This an asynchronous command and the run may continue executing for some time
+This is an asynchronous command and the run may continue executing for some time
 before being terminated.
 
 ## Viewing logs
@@ -313,9 +313,9 @@ the run's containers.
 `--follow` streams logs to standard output as they are written until the
 run completes.
 
-`--since` only shows logs after a a given time.
+`--since` only shows logs after a given time.
 
-`--tail` only shows new last N log lines.
+`--tail` only shows the last N log lines.
 
 `--timestamps` prefixes each line with its timestamp.