The CHIP Tool (chip-tool) is a Matter controller implementation that allows to
commission a Matter device into the network and to communicate with it using
Matter messages, which may encode Data Model actions, such as cluster commands.
The tool also provides other utilities specific to Matter, such as parsing of the setup payload or performing discovery actions.
You can find source files of the CHIP Tool in the examples/chip-tool
directory.
Note: The CHIP Tool caches the configuration state in the
/tmp/chip_tool_config.inifile. Deleting this and other.inifiles in the/tmpdirectory can sometimes resolve issues related to stale configuration.
Before you can use the CHIP Tool, you must compile it from source on Linux (amd64/aarch64) or macOS. If you want to run it on Raspberry Pi, it must use a 64-bit OS.
Note: To ensure compatibility, always build the CHIP Tool and the Matter device from the same revision of the
connectedhomeiprepository.
To build and run the CHIP Tool:
-
Install all required packages for Matter and prepare the source code and the build system. Read the Building Matter guide for instructions.
-
Open a command prompt in the
connectedhomeipdirectory. -
Run the following command:
./scripts/examples/gn_build_example.sh examples/chip-tool BUILD_PATHIn this command,
BUILD_PATHspecifies where the target binaries are to be placed.
To check if the CHIP Tool runs correctly, execute the following command from the
BUILD_PATH directory:
$ ./chip-tool
As a result, the CHIP Tool starts in the default single-command mode and prints all available commands. These are called clusters in this context, but not all listed commands correspond to the clusters in the Data Model (for example, pairing or discover commands). Each listed command can however become the root of the new more complex command by appending it with sub-commands. Examples of specific commands and their use cases are described in the Supported commands and options section.
The CHIP Tool can run in one of the following modes:
In this mode, the CHIP Tool will exit with a timeout error if any single command does not complete within a certain timeout period.
The timeout error will look similar to the following one:
[1650992689511] [32397:1415601] CHIP: [TOO] Run command failure: ../../../examples/chip-tool/commands/common/CHIPCommand.cpp:392: CHIP Error 0x00000032: Timeout
Moreover, when using the single-command mode, the CHIP Tool will establish a new CASE session with every command sent.
This timeout can be modified for any command execution by supplying the optional
--timeout parameter, which takes a value in seconds, with the maximum being
65535 seconds.
Example of command:
$ ./chip-tool otasoftwareupdaterequestor subscribe-event state-transition 5 10 0x1234567890 0 --timeout 65535
In this mode, a command will terminate with an error if it does not complete within the timeout period. However, the CHIP Tool will not be terminated and it will not terminate processes that previous commands have started. Moreover, when using the interactive mode, the CHIP Tool will establish a new CASE session only when there is no session available yet. On the following commands, it will use the existing session.
For commands such as event subscriptions that need to run for an extended period of time, the CHIP Tool can be started in interactive mode first before running the command.
To start the interactive mode, run the following command:
$ ./chip-tool interactive start
In this mode, you can subscribe to events or attributes. For detailed steps, see Subscribing to events or attributes.
The following steps depend on the application clusters that you implemented on the device.
The steps are using the Matter Lighting Application Example with the Bluetooth LE commissioning method support. You can use other Matter examples and still follow this procedure. If you use a different example, the Step 7 may vary depending on the clusters implemented in your application.
Build and program the device with the Matter device firmware by following the Matter Lighting Application Example documentation.
Some examples are configured to advertise automatically on boot. Other examples require physical trigger, for example pushing a button. Follow the documentation of the Matter device example for the chosen platform to learn how Bluetooth LE advertising is enabled for the given example.
To follow the next steps, the IP network must be up and running. For instance, the Thread network can be established using OpenThread Border Router.
You must provide the CHIP Tool with network credentials that will be used in the device commissioning procedure to configure the device with a network interface, such as Thread or Wi-Fi.
The Matter specification does not define the preferred way of how the network credentials are to be obtained by controller. In this guide, we are going to provide steps for obtaining Thread and Wi-Fi network credentials.
Fetch and store the current Active Operational Dataset from the Thread Border Router. This step may vary depending on the Thread Border Router implementation.
If you are using OpenThread Border Router (OTBR), retrieve this information using one of the following commands:
-
For OTBR running in Docker:
sudo docker exec -it otbr sh -c "sudo ot-ctl dataset active -x" 0e080000000000010000000300001335060004001fffe002084fe76e9a8b5edaf50708fde46f999f0698e20510d47f5027a414ffeebaefa92285cc84fa030f4f70656e5468726561642d653439630102e49c0410b92f8c7fbb4f9f3e08492ee3915fbd2f0c0402a0fff8 Done -
For OTBR native installation:
sudo ot-ctl dataset active -x 0e080000000000010000000300001335060004001fffe002084fe76e9a8b5edaf50708fde46f999f0698e20510d47f5027a414ffeebaefa92285cc84fa030f4f70656e5468726561642d653439630102e49c0410b92f8c7fbb4f9f3e08492ee3915fbd2f0c0402a0fff8 Done
For Thread, you might also use a different out-of-band method to fetch the network credentials.
You must get the following Wi-Fi network credentials to commission the Matter device to the Wi-Fi network:
- Wi-Fi SSID
- Wi-Fi password
The steps required to determine the SSID and password may vary depending on the setup. For instance, you might need to contact your local Wi-Fi network administrator.
Matter uses the following values:
- Discriminator - A 12-bit value used to discern between multiple commissionable device advertisements.
- Setup PIN code - A 27-bit value used to authenticate the device.
You can find these values in the logging terminal of the device (for instance UART) when the device boots up. For example:
I: 254 [DL]Device Configuration:
I: 257 [DL] Serial Number: TEST_SN
I: 260 [DL] Vendor Id: 65521 (0xFFF1)
I: 263 [DL] Product Id: 32768 (0x8000)
I: 267 [DL] Hardware Version: 1
I: 270 [DL] Setup Pin Code: 20202021
I: 273 [DL] Setup Discriminator: 3840 (0xF00)
I: 278 [DL] Manufacturing Date: (not set)
I: 281 [DL] Device Type: 65535 (0xFFFF)
In this printout, the discriminator is 3840 (0xF00) and the setup PIN code is
20202021.
Before communicating with the Matter device, first it must join an existing IP network.
Matter devices can use different commissioning channels:
- Devices that are not yet connected to the target IP network use Bluetooth LE as the commissioning channel.
- Devices that have already joined an IP network only need to use the IP protocol for commissioning to the Matter network.
In this case, your device can join the existing IP network over Bluetooth LE and then be commissioned into a Matter network.
Different scenarios are available for Thread and Wi-Fi networks, as described in the following subsections.
After connecting the device over Bluetooth LE, the controller prints the following log:
Secure Session to Device Established
This log message means that the PASE (Password-Authenticated Session Establishment) session using SPAKE2+ protocol has been established.
To commission the device to the existing Thread network, use the following command pattern:
$ ./chip-tool pairing ble-thread <node_id> hex:<operational_dataset> <pin_code> <discriminator>
In this command:
- <node_id> is the user-defined ID of the node being commissioned.
- <operational_dataset> is the Operational Dataset determined in the step 4.
- <pin_code> and <discriminator> are device-specific keys determined in the step 5.
To commission the device to the existing Wi-Fi network, use the following command pattern:
$ ./chip-tool pairing ble-wifi <node_id> <ssid> <password> <pin_code> <discriminator>
In this command:
- <node_id> is the user-defined ID of the node being commissioned.
- <ssid> and <password> are credentials determined in the step 3.
- <pin_code> and <discriminator> are device-specific keys determined in the step 5.
If you prefer the hexadecimal format, use the hex: prefix. For example:
$ ./chip-tool pairing ble-wifi <node_id> hex:<ssid> hex:<password> <pin_code> <discriminator>
Note: The <node_id> can be provided as a hexadecimal value with the
0xprefix.
This option is available when the Matter device is already present in an IP network, but it has not been commissioned to a Matter network yet.
To commission the device, you can use either the setup PIN code or the setup PIN code and the discriminator, both of which you obtained in the step 5. Alternatively, you can also use a QR code payload.
To discover devices and try to pair with one of them using the provided setup code, use the following command pattern:
$ ./chip-tool pairing onnetwork <node_id> <pin_code>
The command keeps trying devices until pairing with one of them succeeds or until it runs out of pairing possibilities. In this command:
- <node_id> is the user-defined ID of the node being commissioned.
- <pin_code> is device specific setup PIN code determined in the step 5 and used to discover the device.
To discover devices with a long discriminator and try to pair with one of them using the provided setup code, use the following command pattern:
$ ./chip-tool pairing onnetwork-long <node_id> <pin_code> <discriminator>
The command keeps trying devices until pairing with one of them succeeds or until it runs out of pairing possibilities. In this command:
- <node_id> is the user-defined ID of the node being commissioned.
- <pin_code> and <discriminator> are device specific keys determined in the step 5.
Matter devices log the QR code payload and manual pairing code when they boot.
To discover devices based on the given QR code payload or manual pairing code and try to pair with one of them, use the following command pattern:
$ ./chip-tool pairing code <node_id> <qrcode_payload-or-manual_code>
The command keeps trying devices until pairing with one of them succeeds or until it runs out of pairing possibilities. In this command:
- <node_id> is the user-defined ID of the node being commissioned.
- <qrcode_payload-or-manual_code> is the QR code payload ID, for example
MT:Y.K9042C00KA0648G00, or a manual pairing code like749701123365521327694.
In case commissioning needs to be retested, the following command removes the device with the given node ID from the list of commissioned Matter devices:
$ ./chip-tool pairing unpair <node_id>
In this command, <node_id> is the user-defined ID of the node which is going to be forgotten by the CHIP Tool.
Having completed all previous steps, you have the Matter device successfully commissioned to the network. You can now test the device by interacting with Data Model clusters.
In case of the Matter Lighting Application Example we referenced in step 1, the
application implements the On/Off and Level Control clusters. This means that
you can test it by toggling the bulb (using the onoff cluster commands) or by
manipulating its brightness (using the levelcontrol cluster commands):
-
Use the following command pattern to toggle the OnOff attribute state (e.g. visualized by the LED state):
$ ./chip-tool onoff toggle <node_id> <endpoint_id>In this command:
- <node_id> is the user-defined ID of the commissioned node.
- <endpoint_id> is the ID of the endpoint with OnOff cluster implemented.
-
Use the following command pattern to change the value of the CurrentLevel attribute (e.g. visualized by the LED brightness):
$ ./chip-tool levelcontrol move-to-level <level> <transition_time> <option_mask> <option_override> <node_id> <endpoint_id>In this command:
- <level> is the brightness level encoded between
0and254, unless a custom range is configured in the cluster. - <transition_time> is the transition time.
- <option_mask> is the option mask.
- <option_override> is the option override.
- <node_id> is the user-defined ID of the commissioned node.
- <endpoint_id> is the ID of the endpoint with LevelControl cluster implemented.
- <level> is the brightness level encoded between
Every Matter device supports the Basic Information cluster, which maintains the collection of attributes that a controller can obtain from a device. These attributes can include the vendor name, the product name, or the software version.
Use the CHIP Tool's read command on the basicinformation cluster to read
those values from the device:
$ ./chip-tool basicinformation read vendor-name <node_id> <endpoint_id>
$ ./chip-tool basicinformation read product-name <node_id> <endpoint_id>
$ ./chip-tool basicinformation read software-version <node_id> <endpoint_id>
In these commands:
- <node_id> is the user-defined ID of the commissioned node.
- <endpoint_id> is the ID of the endpoint with Basic Information cluster implemented.
You can also use the following command to list all available commands for the Basic Information cluster:
$ ./chip-tool basicinformation
This section contains a general list of various CHIP Tool commands and options, not limited to commissioning procedure and cluster interaction.
To print all clusters supported by the CHIP Tool, run the following command:
$ ./chip-tool
Example of output:
[1647346057.900626][394605:394605] CHIP:TOO: Missing cluster name
Usage:
./chip-tool cluster_name command_name [param1 param2 ...]
+-------------------------------------------------------------------------------------+
| Clusters: |
+-------------------------------------------------------------------------------------+
| * accesscontrol |
| * accountlogin |
| * administratorcommissioning |
| * alarms |
| * any |
| * appliancecontrol |
| * applianceeventsandalert |
| * applianceidentification |
| * appliancestatistics |
| * applicationbasic |
To print the list of commands supported by a specific cluster, use the following command pattern:
$ ./chip-tool <cluster_name>
In this command:
- <cluster_name> is one of the available clusters (listed with
chip-tool).
Example of command:
$ ./chip-tool onoff
Example of output:
[1647417645.182824][404411:404411] CHIP:TOO: Missing command name
Usage:
./chip-tool onoff command_name [param1 param2 ...]
+-------------------------------------------------------------------------------------+
| Commands: |
+-------------------------------------------------------------------------------------+
| * command-by-id |
| * off |
| * on |
| * toggle |
| * off-with-effect |
| * on-with-recall-global-scene |
| * on-with-timed-off |
| * read-by-id |
| * read |
| * write-by-id |
| * write |
| * subscribe-by-id |
| * subscribe |
| * read-event-by-id |
| * subscribe-event-by-id |
+-------------------------------------------------------------------------------------+
[1647417645.183836][404411:404411] CHIP:TOO: Run command failure: ../../examples/chip-tool/commands/common/Commands.cpp:84: Error 0x0000002F
To get the list of attributes for a specific cluster, use the following command pattern:
$ ./chip-tool <cluster_name> read
In this command:
- <cluster_name> is one of the available clusters (listed with
chip-tool).
Example of command:
$ ./chip-tool onoff read
Example of output:
[1647417857.913942][404444:404444] CHIP:TOO: Missing attribute name
Usage:
./chip-tool onoff read attribute-name [param1 param2 ...]
+-------------------------------------------------------------------------------------+
| Attributes: |
+-------------------------------------------------------------------------------------+
| * on-off |
| * global-scene-control |
| * on-time |
| * off-wait-time |
| * start-up-on-off |
| * server-generated-command-list |
| * client-generated-command-list |
| * attribute-list |
| * feature-map |
| * cluster-revision |
+-------------------------------------------------------------------------------------+
[1647417857.914110][404444:404444] CHIP:TOO: Run command failure: ../../examples/chip-tool/commands/common/Commands.cpp:120: Error 0x0000002FTo get the list of parameters for a specific command, use the following command pattern:
$ ./chip-tool <cluster_name> <target_command>
In this command:
- <cluster_name> is one of the available clusters (listed with
chip-tool). - <target_command> is one of the target command names.
Example of command:
$ ./chip-tool onoff on
Example of output:
[1647417976.556313][404456:404456] CHIP:TOO: InitArgs: Wrong arguments number: 0 instead of 2
Usage:
./chip-tool onoff on node-id/group-id endpoint-id-ignored-for-group-commands [--paa-trust-store-path] [--commissioner-name] [--trace_file] [--trace_log] [--ble-adapter] [--timedInteractionTimeoutMs] [--suppressResponse]
[1647417976.556362][404456:404456] CHIP:TOO: Run command failure: ../../examples/chip-tool/commands/common/Commands.cpp:135: Error 0x0000002F
This section lists selected options that can be used to configure the input command.
To choose the Bluetooth adapter used by the CHIP Tool, use the following command pattern:
--ble-adapter <id>
In this command:
- <id> is the ID of HCI device.
Example of usage:
$ ./chip-tool pairing ble-thread 1 hex:0e080000000000010000000300001335060004001fffe002084fe76e9a8b5edaf50708fde46f999f0698e20510d47f5027a414ffeebaefa92285cc84fa030f4f70656e5468726561642d653439630102e49c0410b92f8c7fbb4f9f3e08492ee3915fbd2f0c0402a0fff8 20202021 3840 --ble-adapter 0
Message tracing allows capturing CHIP Tool secure messages that can be used for test automation. The tracing uses several types of flags that control where the traces should go.
The following flags are available:
-
Trace file flag:
--trace_file <filename>Here, <filename> is the name of the file where the trace data is stored. It can be appended to a command in the following way:
$ ./chip-tool pairing <pairing_options> --trace_file <filename> -
Trace log flag:
--trace_log <onoff>Here, <onoff> is a
[0/1]flag, which when set to1prints the trace data with automation logs to the console.
All CHIP Tool commands can be used together with the following commissioner-related flags:
--commissioner-name--commissioner-nodeid
These flags let you control which fabric and node ID are used for the CHIP Tool when interacting with devices. They are especially useful in scenarios where you are working with more than one fabric, but you can also use them with a single CHIP Tool node identity.
This flag lets you control which fabric is used by selecting a specific fabric commissioner.
As per the CHIP Tool implementation, the commissioners are required to have the
following names: alpha for the first one, beta for the second one, gamma
for the third one, 4 for the fourth one, 5 for the fifth one, and so on.
If you don't use this flag, the CHIP Tool assumes that the command is meant for
the alpha commissioner and hence for the fabric associated with this
commissioner.
Example of commands:
$ ./chip-tool any subscribe-by-id '0x0028,0x0028,0x0101,0x0028,0x0028,0x0028' '5,6,0,1,2,4' 100 1000 1 '0,0,1,0,0,0' --keepSubscriptions true
$ ./chip-tool any subscribe-by-id '0x0028,0x0028,0x0101,0x0028,0x0028,0x0028' '5,6,0,1,2,4' 100 1000 2 '0,0,1,0,0,0' --keepSubscriptions true --commissioner-name beta
$ ./chip-tool any subscribe-by-id '0x0028,0x0028,0x0101,0x0028,0x0028,0x0028' '5,6,0,1,2,4' 100 1000 3 '0,0,1,0,0,0' --keepSubscriptions true --commissioner-name gamma
This flag lets you select the node ID to use on the fabric specified with the
--commissioner-name flag.
If you don't use this flag, the CHIP Tool assumes that the command is sent with
the ID value that the CHIP Tool has in storage. If there's none, the CHIP Tool
sends the command with the default fallback node ID 112233.
Note: If the device has been already commissioned with a specific
--commissioner-nodeid, you must always provide the--commissioner-nodeidflag with the CHIP Tool commands or update the Access Control List (ACL) on the device. Otherwise, the default fallback node ID112233is used and the communication will fail.
Example of commands:
$ ./chip-tool pairing code-thread 1 hex:000030000150208562618342348532605109bd31cda6908667addca8789211addac0102c4a9 34970112332 --commissioner-name alpha --commissioner-nodeid 999999
$ ./chip-tool basicinformation read vendor-id --commissioner-name alpha --commissioner-nodeid 999999
The CHIP Tool allows to run a set of tests, already compiled in the tool, against a paired Matter device.
-
To get the list of available tests, run the following command:
$ ./chip-tool tests -
To execute a particular test against the paired device, use the following command pattern:
$ ./chip-tool tests <test_name>In this command:
- <test_name> is the name of the particular test.
See the Examples section for an example of how to run a test from the test suite.
The CHIP Tool offers a utility for parsing the Matter onboarding setup payload and printing it in a readable form. For example, the payload can be printed on the device console during boot.
To parse the setup code, use the payload command with the
parse-setup-payload sub-command, as in the following command pattern:
$ ./chip-tool payload parse-setup-payload <payload>
Here, <payload> is the ID of the payload to be parsed.
Examples of commands:
-
Setup QR code payload:
$ ./chip-tool payload parse-setup-payload MT:6FCJ142C00KA0648G00 -
Manual pairing code:
$ ./chip-tool payload parse-setup-payload 34970112332
To parse additional data payload, use the following command pattern:
$ ./chip-tool payload parse-additional-data-payload <payload>
In this command:
- <payload> is the ID of the payload with additional data to be parsed.
The discover command can be used to resolve node ID and discover available
Matter devices.
Use the following command to print the available sub-commands of the discover
command:
$ ./chip-tool discover
To resolve the DNS-SD name corresponding with the given Node ID and update the address of the node in the device controller, use the following command pattern:
$ ./chip-tool discover resolve <node_id> <fabric_id>
In this command:
- <node_id> is the ID of node to be resolved.
- <fabric_id> is the ID of the Matter fabric to which the node belongs.
To discover all Matter commissionables available in the operation area, run the following command:
$ ./chip-tool discover commissionables
To discover all Matter commissioners available in the operation area, run the following command:
$ ./chip-tool discover commissioners
The pairing command supports different methods for Matter device commissioning
procedure. The recommended methods are the following:
code-thread- For Thread commissioning.code-wifi- For Wi-Fi commissioning.code- For commissioning the device when it is already present in an IP network.
Alternatively, you can also use the following methods described in the Using CHIP Tool for Matter device testing section:
ble-thread- For Thread commissioning; described under Commissioning into a Thread network over Bluetooth LE.ble-wifi- For Wi-Fi commissioning; described under Commissioning into a Wi-Fi network over Bluetooth LEonnetwork- For commissioning the device when it is already present in an IP network; described under Commissioning with setup PIN code
To list all pairing sub-commands and commissioning methods, run the following
command:
$ ./chip-tool pairing
Example of commands:
The following command commissions the Thread device with the node ID 1 to the
Matter fabric. The hex:... parameter is the operational dataset that contains
information about the Thread network to which the device is going to be
commissioned. The onboarding dataset payload 34970112332 (short manual pairing
code) is used to discover and commission the device.
$ ./chip-tool pairing code-thread 1 hex:000030000150208562618342348532605109bd31cda6908667addca8789211addac0102c4a9 34970112332
The following command commissions the Wi-Fi device with the node ID 1 to the
Matter fabric The SSID wifi_test and the password admin123 is the required
information about the Wi-Fi network to which the device is going to be
commissioned. The onboarding dataset payload 34970112332 (short manual pairing
code) is used to discover and commission the device.
$ ./chip-tool pairing code-wifi 1 wifi_test admin123 34970112332
The following command commissions the device with the node ID 1 to the Matter
fabric. The onboarding dataset payload MT:8IXS142C00KA0648G00 (QR code
payload) is used to discover and commission the device.
$ ./chip-tool pairing code 1 MT:8IXS142C00KA0648G00
The pairing commissioning command can be run with several flags that allow you
to modify attestation-related settings:
-
--paa-trust-store-path- Use to provide the path to the directory that contains the information about Product Attestation Authority (PAA) certificates. The path can be absolute or relative to the current working directory. With this flag, the CHIP Tool looks for the PAA certificate that matches the PAI and the DAC certificates programmed on the device. Without this flag, the CHIP Tool uses the built-in test PAA certificate. -
--cd-trust-store-path- Use to provide the path to the directory containing the key that is used to validate the device's Certification Declaration. The path can be absolute or relative to the current working directory. With this flag, the CHIP tool looks for additional public keys, in addition to the well-known built-in public keys (built-in public keyssrc/credentials/attestation_verifier/DefaultDeviceAttestationVerifier.cpp), to be used to validate Certification Declaration signatures.
-
--only-allow-trusted-cd-keys- Use to only allow the keys from--cd-trust-store-pathand not the built-in test key. If the flag is not provided or it is provided with the valuefalse, untrusted CD verifying keys are allowed. If it is provided with the valuetrue(--only-allow-trusted-cd-keys true), test keys are disallowed and CD signed with the test key will not be accepted. -
--bypass-attestation-verifier- Use to bypass the attestation verifier. If the flag is not provided or it is provided with the valuefalse, the attestation verifier is not bypassed. If it is provided with the valuetrue(--bypass-attestation-verifier true), the commissioning will continue in case of the attestation verification failure. The failure can be caused by errors in Certification Declaration, PAA or PAI certificates, or in the Device Attestation Certificate. This option can be helpful if you want to quickly commission a device with PAI and DAC certificates based on an unknown PAA and/or with a Certification Declaration signed by an unknown key.
As mentioned in the
Using CHIP Tool for Matter device testing
section, executing the chip-tool command with a particular cluster name lists
all operations supported for this cluster, as in the following command pattern:
$ ./chip-tool <cluster_name>
Example of command:
$ ./chip-tool binding
Example of output:
[1647502596.396184][411686:411686] CHIP:TOO: Missing command name
Usage:
./chip-tool binding command_name [param1 param2 ...]
+-------------------------------------------------------------------------------------+
| Commands: |
+-------------------------------------------------------------------------------------+
| * command-by-id |
| * read-by-id |
| * read |
| * write-by-id |
| * write |
| * subscribe-by-id |
| * subscribe |
| * read-event-by-id |
| * subscribe-event-by-id |
+-------------------------------------------------------------------------------------+
[1647502596.396299][411686:411686] CHIP:TOO: Run command failure: ../../examples/chip-tool/commands/common/Commands.cpp:84: Error 0x0000002FAccording to this list, the binding cluster supports operations such as read
or write. Attributes from that cluster can also be subscribed by the controller,
which means that the CHIP Tool will receive notifications, for instance when the
attribute value is changed or a specific event happens.
This section lists examples of the CHIP Tool commands dedicated to specific use cases.
The Access Control List concept allows to govern all Data Model interactions (such as read attribute, write attribute, invoke command). For more information about ACL, see Access Control Guide.
To write ACL to the accesscontrol cluster, use the following command pattern:
$ ./chip-tool accesscontrol write acl <acl_data> <node_id> <endpoint_id>
In this command:
- <acl_data> is the ACL data formatted as a JSON array.
- <node_id> is the ID of the node that is going to receive ACL.
- <endpoint_id> is the ID of the endpoint on which the
accesscontrolcluster is implemented.
Binding describes a relationship between the device that contains the binding cluster and the end device. The proper ACL must be added to allow the end device to receive commands from the bonded device. After the binding process, the bonded device contains information about connected device, such as IPv6 address and the route to the endpoint in the Matter network.
To add a binding table to the binding cluster, use the following command
pattern:
$ ./chip-tool binding write binding <binding_data> <node_id> <endpoint_id>
In this command:
- <binding_data> is the binding data formatted as a JSON array.
- <node_id> is the ID of the node that is going to receive the binding.
- <endpoint_id> is the ID of the endpoint on which the
bindingcluster is implemented.
Complete the following steps to run one test from the test suite:
- Clean the initialization of state using the following command:
rm -fr /tmp/chip_* - In a shell window, start the DUT device:
./out/debug/standalone/chip-all-clusters-app - In a second shell window, pair the DUT with the CHIP Tool:
./out/debug/standalone/chip-tool pairing onnetwork 333221 20202021 - Run the test with the following command:
./out/debug/standalone/chip-tool tests TestCluster --nodeId 333221
Read the CHIP Test Suits page for more information about how the test suite is structured.
Multi-admin feature allows you to join Matter device to several Matter fabrics and have several different Matter administrators administer it.
Complete the steps mentioned in the following sections.
Commission the Matter device to first fabric following the Using CHIP Tool for Matter device testing section.
Make sure the administrator from the first fabric opens the commissioning window for a new administrator from another fabric.
Open the commissioning window on the paired Matter device by using the following command pattern:
$ ./chip-tool pairing open-commissioning-window <node_id> <option> <window_timeout> <iteration> <discriminator>
In this command:
- <node_id> is the ID of the node that should open commissioning window.
- <option> is equal to 1 for Enhanced Commissioning Method and 0 for Basic Commissioning Method.
- <window_timeout> is time in seconds, before the commissioning window closes.
- <iteration> is number of PBKDF iterations to use to derive the PAKE verifier.
- <discriminator> is device specific discriminator determined during commissioning.
Note: The <iteration> and <discriminator> values are ignored if the <option> is set to 0.
Example of command:
$ ./chip-tool pairing open-commissioning-window 1 1 300 1000 2365
Write down the manual pairing code or the QR code payload printed in the command output, as it will be required by the second Matter admin to join Matter device to its fabric.
Example of output:
[1663675289.149337][56387:56392] CHIP:DMG: Received Command Response Status for Endpoint=0 Cluster=0x0000_003C Command=0x0000_0000 Status=0x0
[1663675289.149356][56387:56392] CHIP:CTL: Successfully opened pairing window on the device
[1663675289.149409][56387:56392] CHIP:CTL: Manual pairing code: [36281602573]
[1663675289.149445][56387:56392] CHIP:CTL: SetupQRCode: [MT:4CT91AFN00YHEE7E300]
Complete the following steps:
-
Open another instance of the CHIP Tool.
-
In the new instance of the CHIP Tool, commission the Matter device to a new fabric by using the following command pattern:
$ ./chip-tool pairing code <node_id> <payload> --commissioner-name <commissioner_name>In this command:
- <node_id> is the user-defined ID of the node being commissioned. It doesn't need to be the same ID, as for the first fabric.
- <payload> is the the QR code payload or a manual pairing code generated by the first commissioner instance when opened commissioning window
- <commissioner_name> is the name of the second fabric. Valid values are "alpha", "beta", "gamma", and integers greater than or equal to 4. The default if not specified is "alpha".
Example of command:
$ ./chip-tool pairing code 1 36281602573 --commissioner-name beta
After completing the above steps, the Matter device should be able to receive
and answer Matter commands sent in the second fabric. For example, you can use
the following command pattern to toggle the OnOff attribute state on a device
supporting OnOff cluster:
$ ./chip-tool onoff toggle <node_id> <endpoint_id> --commissioner-name <commissioner_name>
In this command:
- <node_id> is the user-defined ID of the commissioned node.
- <endpoint_id> is the ID of the endpoint with OnOff cluster implemented.
- <commissioner_name> is the name of the second fabric. Valid values are "alpha", "beta", "gamma", and integers greater than or equal to 4. The default if not specified is "alpha".
Example of command:
$ ./chip-tool onoff toggle 1 1 --commissioner-name beta
Subscribing to an event or an attribute lets you mirror the state of the event or the attribute as it changes in the Matter network. The list of events or attributes you can subscribe to depends on the chosen cluster.
You can have more than one subscription at any given time and subscribe to more than one attribute or event within one subscription (those attributes or events can come from different clusters). However, you cannot subscribe to both attributes and events as part of a single subscription. In other words, each subscription must be dedicated exclusively to either attributes or events.
For more information about subscriptions, see the Matter specification at chapter 8, section 5 (Subscribe Interaction).
Note: The subscription behavior will be different if you set the subscription to be sent with the parameter
isUrgentset toTrue. See the Matter specification for more information.
The following procedure will use the doorlock cluster as an example. Complete
the following steps:
-
Start the CHIP Tool in interactive mode by running the following command:
$ ./chiptool interactive startAll of the commands that follow will be executed in the interactive mode (
>>>). -
Run the following command to display all the available attributes you can subscribe to for the given
<cluster-name>:>>> <cluster-name> subscribeThe list of all available attributes for the cluster will appears.
Note: Your accessory might not support all of these attributes. You will get an error if the controller sends an unsupported attribute.
For example, for the door lock cluster:
>>> doorlock subscribe
The following list will appear:
+-------------------------------------------------------------------------------------+ | Attributes: | +-------------------------------------------------------------------------------------+ | * lock-state | | * lock-type | | * actuator-enabled | | * door-state | | * door-open-events | | * door-closed-events | | * open-period | | * number-of-total-users-supported | | * number-of-pinusers-supported | | * number-of-rfidusers-supported | | * number-of-week-day-schedules-supported-per-user | | * number-of-year-day-schedules-supported-per-user | | * number-of-holiday-schedules-supported | | * max-pincode-length | | * min-pincode-length | | * max-rfidcode-length | | * min-rfidcode-length | | * credential-rules-support | | * number-of-credentials-supported-per-user | | * language | | * ledsettings | | * auto-relock-time | | * sound-volume | | * operating-mode | | * supported-operating-modes | | * default-configuration-register | | * enable-local-programming | | * enable-one-touch-locking | | * enable-inside-status-led | | * enable-privacy-mode-button | | * local-programming-features | | * wrong-code-entry-limit | | * user-code-temporary-disable-time | | * send-pinover-the-air | | * require-pinfor-remote-operation | | * expiring-user-timeout | | * generated-command-list | | * accepted-command-list | | * event-list | | * attribute-list | | * feature-map | | * cluster-revision | +-------------------------------------------------------------------------------------+ -
Add the argument of your choice to the subscription command, using the following pattern:
>>> <cluster-name> subscribe <argument> <min-interval> <max-interval> <node_id> <endpoint_id>
In this command:
- <cluster-name> is the name of the cluster.
- <argument> is the name of the chosen argument.
- <min-interval> specifies the minimum number of seconds that must elapse since the last report for the server to send a new report.
- <max-interval> specifies the number of seconds that must elapse since the last report for the server to send a new report.
- <node-id> is the user-defined ID of the commissioned node.
- <endpoint_id> is the ID of the endpoint where the chosen cluster is implemented.
For example:
>>> doorlock subscribe lock-state 5 10 1 1
After this command is run, the CHIP Tool will check the state of the door lock every time it changes (for example, as a result of a button press or an external ecosystem action) and update it in its own records.
The procedure for subscribing to an event is similar to subscribing to an attribute.
The following procedure will use the doorlock cluster as an example. Complete
the following steps:
-
Start the CHIP Tool in interactive mode by running the following command:
$ ./chiptool interactive startAll of the commands that follow will be executed in the interactive mode (
>>>). -
Run the following command to display all the available events you can subscribe to for the given
<cluster-name>:>>> <cluster-name> subscribe-eventThe list of all available events for the cluster will appears.
Note: Your accessory might not support all of these events. You will get an error if the controller sends an unsupported event.
For example, for the door lock cluster:
>>> doorlock subscribe-event
The following list will appear:
+-------------------------------------------------------------------------------------+ | Events: | +-------------------------------------------------------------------------------------+ | * door-lock-alarm | | * door-state-change | | * lock-operation | | * lock-operation-error | | * lock-user-change | +-------------------------------------------------------------------------------------+ -
Add the event of your choice to the subscription command, using the following pattern:
>>> <cluster-name> subscribe-event <event-name> <min-interval> <max-interval> <node_id> <endpoint_id>
In this command:
- <cluster-name> is the name of the cluster.
- <event-name> is the name of the chosen event.
- <min-interval> specifies the minimum number of seconds that must elapse since the last report for the server to send a new report.
- <max-interval> specifies the number of seconds that must elapse since the last report for the server to send a new report.
- <node_id> is the user-defined ID of the commissioned node.
- <endpoint_id> is the ID of the endpoint where the chosen cluster is implemented.
For example:
>>> doorlock subscribe-event door-lock-alarm 5 10 1 1
After this command is run, the CHIP Tool will check the state of the door lock every time it changes (for example, as a result of a button press or an external ecosystem action) and update it in its own records.
You can also use the following commands instead of subscribe to subscribe
using the attribute ID or the event ID:
subscribe-by-idsubscribe-event-by-id
The steps are the same as for the subscribe or subscribe-event commands.
The CHIP Tool supports command wildcards for parameter values for clusters,
attributes or events, or endpoints, or any combination of these. With the
wildcards, you can for example read all attributes for the cluster 0x101 on a
specific endpoint with a specific node ID on all devices in the Matter network.
This allows you to parse and gather cluster information faster and more
efficiently.
The following wildcards are available:
- For all attributes:
0xFFFFFFFF - For all clusters:
0xFFFFFFFF - For all endpoints:
0xFFFF
You can combine these wildcards within a single command. Wildcards can be used in both single-command and interactive modes.
You can use the following command pattern:
$ ./chip-tool <cluster_name> <command> <attribute_event_name> <node_id> <endpoint_id>
In this command:
-
<cluster-name> is the name of the cluster.
-
<command> is the name of the command supported by wildcards:
+-------------------------------------------------------------------------------------+ | Commands: | +-------------------------------------------------------------------------------------+ | * read | | * read-by-id | | * subscribe | | * subscribe-by-id | +-------------------------------------------------------------------------------------+ -
<attribute_event_name> is the name of the chosen attribute or event.
-
<node_id> is the user-defined ID of the commissioned node.
-
<endpoint_id> is the ID of the endpoint where the chosen cluster is implemented.
Examples of commands:
-
To read all attributes (wildcard
0xFFFFFFFF) from the clusterdoorlockfor the node with ID1and on the endpoint1, run the following command:$ ./chip-tool doorlock read-by-id 0xFFFFFFFF 1 1 -
To read the
lock-stateattribute from the clusterdoorlockfor the node with ID1and on all endpoints (wildcard0xFFFF), run the following command:$ ./chip-tool doorlock read lock-state 1 0xFFFF -
To read all attributes (wildcard
0xFFFFFFFF) from the clusterdoorlockfor the node with ID1and on all endpoints (wildcard0xFFFF), run the following command:$ ./chip-tool doorlock read-by-id 0xFFFFFFFF 1 0xFFFF
Using the any command lets you use wildcards also for the cluster names. The
any command can be combined with the following commands:
+-------------------------------------------------------------------------------------+
| Commands: |
+-------------------------------------------------------------------------------------+
| * command-by-id |
| * read-by-id |
| * write-by-id |
| * subscribe-by-id |
| * read-event-by-id |
| * subscribe-event-by-id |
| * read-all |
| * subscribe-all |
+-------------------------------------------------------------------------------------+
As a result, you can use the following command pattern:
$ ./chip-tool any <command_name> [parameters of the <command_name>]
In this command:
- <command_name> is one of the commands supported for the
anycommand, as listed above. - [parameters of the <command_name>] are the parameters required by <command_name>. You can check them by running the command without any parameters.
Example of command pattern for read-by-id:
$ ./chip-tool any read-by-id <cluster-ids> <attribute-ids> <destination-id> <endpoint-ids>
Examples of commands:
-
To read the
0x0attribute (lock state) on the cluster0x101(doorlock) for the node with ID1and on the endpoint1, run the following command:$ ./chip-tool any read-by-id 0x101 0x0 1 1 -
To read all attributes (wildcard
0xFFFFFFFF) from the cluster0x101(doorlock) for the node with ID1and on the endpoint1, run the following command:$ ./chip-tool any read-by-id 0x101 0xFFFFFFFF 1 1 -
To read all attributes (wildcard
0xFFFFFFFF) on all clusters (wildcard0xFFFFFFFF) for the node with ID1and on the endpoint1, run the following command:./chip-tool any read-by-id 0xFFFFFFFF 0xFFFFFFFF 1 1 -
To read all attributes (wildcard
0xFFFFFFFF) on all clusters (wildcard0xFFFFFFFF) for the node with ID1and on all endpoints (wildcard0xFFFF), run the following command:./chip-tool any read-by-id 0xFFFFFFFF 0xFFFFFFFF 1 0xFFFF
Matter door lock devices can store pools of users and credentials that allow you to configure different access scenarios. Each user and credential in the pool has an index value. Additionally, each user has a list of Occupied Credentials.
By default, each door lock device comes with no user or credential defined, but
it reserves several slots in both pools that can be populated with new users and
credentials, up to the value specified in the the NumberOfTotalUsersSupported
attribute and the NumberOfCredentialsSupportedPerUser attribute, respectively.
All communication between users and credentials happens only using their respective index values. No other information is shared between both pools.
The CHIP Tool lets you add users and credentials on door lock devices and
securely match their indexes to one another. This is an optional feature
available only when working with devices that implement the doorlock cluster.
Note: Users and credentials can only be modified by whoever has the right permissions, as specified in the Access Control List.
To save credentials and users, you need to complete the following steps, described in detail in the following sections:
- Set up a user on the device.
- Assign a credential for the newly created user.
To set up a user on a door lock device with the CHIP Tool, use the following command pattern:
$ ./chip-tool doorlock set-user <operation-type> <user-index> <user-name> <user-unique-id> <user-status> <user-type> <credential-rule> <destination-id> <endpoint-id> --timedInteractionTimeoutMs <ms_value>
In this command:
-
<operation-type> is one of the available types of operation for the user:
Add- This operation sets a new user in the slot at <user-index>.Clear- This operation removes an existing user from the slot at <user-index>.Modify- This operation modifies an existing user at the slot at <user-index>.
-
<user-index> is the index value of the user, between
1and the value of theNumberOfTotalUsersSupportedattribute. Setting the user index to0will cause an error. -
<user-name> is the name of the user, which can have maximum 10 bytes of size. Can be set to
null. -
<user-unique-id> is a 4-byte number that describes the unique user ID. Can be set to
null. -
<user-status> can be set to
nullor to one of the following values:1(OccupiedEnabled) - This status indicates that the given user slot is used and active.3(OccupiedDisabled) - This status indicates that the given user slot is used, but disabled. Unlike0and1, supporting this status is optional.
-
<user-type> is the type of the user, which can have one of the values specified in the Matter Application Clusters specification for the
doorlockcluster (see section "5.2.9.16.UserTypeEnum"). Can be set tonull. -
<credential-rule> is the number of credentials that must be used to unlock the door lock. This parameter can be set to
nullor to one of the following values:0(Single) - One credential type is required to unlock.1(Dual) - Two credential types are required to unlock.2(Triple) - Three credential types are required to unlock.
-
<destination-id> is the node ID of the door lock device.
-
<endpoint-id> is the ID of the endpoint on the door lock device.
-
--timedInteractionTimeoutMsis the duration in milliseconds (<ms_value>) of the time window for receiving a request on the server side. This should allow enough time for receiving the request.
Examples of command:
The following command runs the set-user command that adds (0) a user at the
index 1; the user has the name AAA and the unique ID 6452. The user's
status is set to OccupiedEnabled (1), the user type is set to
UnrestrictedUser (0), the credential rule is set to single (0), the
targeted node ID of the destination door lock device is 1 and the targeted
doorlock cluster's endpoint ID on that device is 1. The
--timedInteractionTimeoutMs has a custom value.
$ ./chip-tool doorlock set-user 0 1 AAA 6452 1 0 0 1 1 --timedInteractionTimeoutMs 1000
The following command mirrors the action of the command above, but it targets an
empty user name (null) and has null for the unique ID. The user status
defaults to OccupiedEnabled, the user type defaults to UnrestrictedUser, and
the credential rule defaults to single.
$ ./chip-tool doorlock set-user 0 1 null null null null null 1 1 --timedInteractionTimeoutMs 1000
For more use cases for this command, see the "5.2.7.34. Set User Command" section in the Matter Application Clusters specification.
Once you have a user created on the door lock device, use the following command pattern to assign a credential to it:
$ ./chip-tool doorlock set-credential <operation-type> <{Credential}> <credential-data> <user-index> <user-status> <user-type> <destination-id> <endpoint-id> --timedInteractionTimeoutMs <ms_value>
In this command:
-
<operation-type> is one of the available types of operation for the credential:
Add- This operation adds a new credential to a user at the slot at <user-index>.Clear- This operation removes an existing credential from the user at the slot at <user-index>.Modify- This operation modifies an existing credential for the user at the slot at <user-index>.
-
<{Credential}> is a JSON object, with the following two fields:
-
"credentialType"is the key field for the type of the credential. It can have one of the following values:0- Programming PIN1- PIN2- RFID3- Fingerprint4- Finger vein
-
"credentialIndex"is the key field for the index of the credential. If"credentialType"is not "Programming PIN","credentialIndex"must be between1and the value of theNumberOfCredentialsSupportedPerUserattribute (see the section 5.2.3.20 of the Matter Application Clusters specification for details).0is required for the Programming PIN. In other cases, setting the credential index to0will cause an error.
-
-
<credential-data> is an octet string parameter with the secret credential data. For example, the PIN code value (
12345in the example below). -
<user-index> is the index of the user that will be associated with the credential. Can be set to
nullto create a new user. -
<user-status> is the status of the user that will be associated with the credential. See the description of this parameter in Set up a user. Can be set to
null. -
<user-type> is the type of the user, which can have one of the values specified in the Matter Application Clusters specification for the
doorlockcluster (see section "5.2.9.16.UserTypeEnum"). Can be set tonull. -
<destination-id> is the node ID of the door lock device.
-
<endpoint-id> is the ID of the endpoint on the door lock device.
-
--timedInteractionTimeoutMsis the duration in milliseconds (<ms_value>) of the time window for receiving a request on the server side. This should allow enough time for receiving the request.
Example of command:
The following command runs the set-credential command that adds (0) a PIN
credential (type 1) at the index 1. The credential data is set to 12345
(PIN code value). This credential is associated with the user at the index 1.
The null parameters for the user status and the user type indicate that the
credentials are added to an existing user. The targeted node ID of the
destination door lock device is 1 and the targeted doorlock cluster's
endpoint ID on that device is 1. The --timedInteractionTimeoutMs has a
custom value.
$ ./chip-tool doorlock set-credential 0 '{ "credentialType": 1, "credentialIndex": 1 }' "12345" 1 null null 1 1 --timedInteractionTimeoutMs 1000
For more use cases for this command, see the "5.2.7.40. Set Credential Command" section in the Matter Application Clusters specification.
After you set up users and credentials on your door lock device, you can use several CHIP Tool commands to interact with them.
All commands reuse the parameters explained earlier in this section. The following command patterns are available:
-
Reading the status of the user:
$ ./chip-tool doorlock get-user <user-index> <destination-id> <endpoint-id>This command returns the status of the user at the specified <user-index> at the specified <destination-id> and <endpoint-id>.
-
Reading the status of the credential:
$ ./chip-tool doorlock get-credential-status <{Credential}> <destination-id> <endpoint-id>This command returns the status of the credential of the specified <{Credential}> at the specified <destination-id> and <endpoint-id>.
-
Cleaning the user:
$ ./chip-tool doorlock clear-user <user-index> <destination-id> <endpoint-id> --timedInteractionTimeoutMs <ms_value>This command cleans the slot containing the specified <user-index> at the specified <destination-id> and <endpoint-id>.
-
Cleaning the credential:
$ ./chip-tool doorlock clear-credential <{Credential}> <destination-id> <endpoint-id> --timedInteractionTimeoutMs <ms_value>This command cleans the slot containing the specified <{Credential}> at the specified <destination-id> and <endpoint-id>.
If you set the <credential-type> to PIN when assigning credentials, you can use the following command patterns to verify if it works and invoke it to open or close the door lock:
-
Verifying the PIN code:
$ ./chip-tool doorlock read require-pinfor-remote-operation <destination-id> <endpoint-id>This command returns either
falseortrue:falseindicates that providing the PIN code is not required to close or open the door lock.trueindicates that the PIN code is required to close or open the door lock.
-
Changing the requirement for the PIN code usage:
$ ./chip-tool doorlock write require-pinfor-remote-operation true <destination-id> <endpoint-id>This command modifies the setting of
require-pinfor-remote-operationtotrue. After you run it, you will have to use the PIN code to lock or unlock the door. -
Closing the door lock with the PIN code:
$ ./chip-tool doorlock lock-door <destination-id> <endpoint-id> --timedInteractionTimeoutMs <ms_value> --PinCode 12345In this command, you need to provide
--PinCodecorresponding to the PIN code you set with <credential-data> (for example12345). -
Opening the door lock with the PIN code:
$ ./chip-tool doorlock unlock-door <destination-id> <endpoint-id> --timedInteractionTimeoutMs <ms_value> --PinCode 12345In this command, you need to provide
--PinCodecorresponding to the PIN code you set with <credential-data> (for example12345).