Get Started on Cloud
BasicsAI AgentsPipelinesActions & ServicesEnvironmentsIntegrationsTargetsTestsDomainsSandboxesTerraformTunnelsYAMLBuddy GitCLIAPIFAQ

Claude Code

View as Markdown
  • PATCH
  • /workspaces
  • /:workspace
  • /projects
  • /:project_name
  • /pipelines
  • /:pipeline_id
  • /actions
  • /:action_id

Update Claude Code action configuration

Warning
You can't change the type of an existing action. To change the type, remove the action and add it again.

Request

REQUIRED SCOPES
PIPELINE_WRITE
URL PARAMETERS
workspacerequiredstring
The human-readable ID of the workspace

Example: "my-company"

project_namerequiredstring
The human-readable ID of the project

Example: "my-project"

pipeline_idrequiredinteger
The ID of the pipeline

Example: 123

action_idrequiredinteger
The ID of the action

Example: 456

BODY PARAMETERS
typerequiredstring
The type of the action

Value: CLAUDE_CODE

after_action_idinteger
The numerical ID of the action, after which this action should be added
namerequiredstring
The name of the action
trigger_timestring enum
Specifies when the action should be executed. ON_SUCCESS actions run after every successful run, also one ended by an EARLY_EXIT action
Allowed enum:
ON_EVERY_EXECUTION,
ON_SUCCESS,
ON_FAILURE,
ON_BACK_TO_SUCCESS,
ON_WARNING,
ON_WAIT_FOR_APPROVE,
ON_TERMINATE
trigger_conditionsTriggerConditionView[]
The list of trigger conditions to meet so that the action can be triggered
run_nextstring enum
Defines whether the action should run in parallel with the next one
Allowed enum:
WAIT_ON_SUCCESS,
IN_SOFT_PARALLEL,
IN_HARD_PARALLEL
run_only_on_first_failureboolean
Defines whether the action should be executed on each failure. Restricted to and required if the trigger_time is ON_FAILURE
variablesEnvironmentVariableView[]
The list of variables you can use the action
disabledboolean
When set to true the action is disabled. By default it is set to false
timeoutinteger
The timeout in seconds
ignore_errorsboolean
If set to true the execution will proceed, mark action as a warning and jump to the next action. Doesn't apply to deployment actions
retry_countinteger
Number of retries if the action fails
retry_intervalinteger
Delay time between auto retries in seconds
loopstring[]
The list of commands that will be executed in a loop
current_revisionstring
The git revision of the last successful run of the action
working_directorystring
The directory in which the pipeline filesystem will be mounted
setup_commandsstring[]
The command that will be executed only on the first run
servicesServiceView[]
The list of services attached to the build environment
main_service_namestring
The hostname of the container in which the action is run. The container will be available under this name in the docker network for services
integrationDockerIntegrationRef
The integration used by the action
promptsLlmBuildPromptView[]
List of prompts to send to the agent. Each prompt has a type (INLINE or FILE) and a value.
modelstring
The Claude model to use (e.g. sonnet, opus, haiku).
skillsstring[]
List of skills to install before the agent runs. Each entry is a source: owner/repo, a full Git/GitHub/GitLab URL (optionally pointing at a single skill via a /tree/<ref>/<subdir> path), or a path relative to the workspace.
mcp_serversobject
Map of MCP servers to connect to, keyed by server name. Tools exposed by declared servers are auto-allowed.
effortstring
Reasoning effort: low, medium, high, xhigh, or max (default high).
claude_argsstring
Additional CLI arguments passed to Claude Code.
dangerously_skip_permissionsboolean
If true, skip permission prompts (runs with --dangerously-skip-permissions).

Response

RESPONSE BODY
typestring
The type of the action

Value: CLAUDE_CODE

urlread-onlystring
API endpoint to GET this object
html_urlread-onlystring
Web URL to view this object in Buddy.works
idinteger
The ID of the action
namestring
The name of the action
current_revisionstring
The git revision of the last successful run of the action
trigger_timestring enum
Specifies when the action should be executed. ON_SUCCESS actions run after every successful run, also one ended by an EARLY_EXIT action
Allowed enum:
ON_EVERY_EXECUTION,
ON_SUCCESS,
ON_FAILURE,
ON_BACK_TO_SUCCESS,
ON_WARNING,
ON_WAIT_FOR_APPROVE,
ON_TERMINATE
trigger_conditionsTriggerConditionView[]
The list of trigger conditions to meet so that the action can be triggered
run_nextstring enum
Defines whether the action should run in parallel with the next one
Allowed enum:
WAIT_ON_SUCCESS,
IN_SOFT_PARALLEL,
IN_HARD_PARALLEL
run_only_on_first_failureboolean
Defines whether the action should be executed on each failure. Restricted to and required if the trigger_time is ON_FAILURE
variablesEnvironmentVariableView[]
The list of variables you can use the action
disabledboolean
When set to true the action is disabled. By default it is set to false
timeoutinteger
The timeout in seconds
ignore_errorsboolean
If set to true the execution will proceed, mark action as a warning and jump to the next action. Doesn't apply to deployment actions
retry_countinteger
Number of retries if the action fails
retry_intervalinteger
Delay time between auto retries in seconds
last_execution_statusstring enum
The status of the last run of the action
Allowed enum:
INPROGRESS,
ENQUEUED,
TERMINATED,
SUCCESSFUL,
FAILED,
INITIAL,
NOT_EXECUTED,
SKIPPED,
TERMINATING,
WAITING_FOR_APPLY,
WAITING_FOR_VARIABLES,
WAITING_FOR_SETTABLE_VARIABLES,
WAITING_FOR_VT_SESSION
pipelineShortPipelineView
Short representation of a pipeline
permissionsPermissionsView
Access permissions configuration
loopstring[]
The list of commands that will be executed in a loop
working_directorystring
The directory in which the pipeline filesystem will be mounted
setup_commandsstring[]
The command that will be executed only on the first run
servicesServiceView[]
The list of services attached to the build environment
main_service_namestring
The hostname of the container in which the action is run. The container will be available under this name in the docker network for services
promptsLlmBuildPromptView[]
List of prompts to send to the agent. Each prompt has a type (INLINE or FILE) and a value.
modelstring
The Claude model to use (e.g. sonnet, opus, haiku).
skillsstring[]
List of skills to install before the agent runs. Each entry is a source: owner/repo, a full Git/GitHub/GitLab URL (optionally pointing at a single skill via a /tree/<ref>/<subdir> path), or a path relative to the workspace.
mcp_serversobject
Map of MCP servers to connect to, keyed by server name. Tools exposed by declared servers are auto-allowed.
integrationIntegrationIdentifierView
The integration used by the action
effortstring
Reasoning effort: low, medium, high, xhigh, or max (default high).
claude_argsstring
Additional CLI arguments passed to Claude Code.
dangerously_skip_permissionsboolean
If true, skip permission prompts (runs with --dangerously-skip-permissions).

Last modified on Nov 10, 2026

Previous pageGet actionNext pageAdd action

Request example

curl -X PATCH "https://api.buddy.works/workspaces/:workspace/projects/:project_name/pipelines/:pipeline_id/actions/:action_id" \
  -H "Authorization: Bearer <YOUR-TOKEN>" \
  -H "Content-Type: application/json" \
  -d '{
  "name": "Claude Code",
  "type": "CLAUDE_CODE",
  "trigger_time": "ON_EVERY_EXECUTION",
  "prompts": [
    {
      "type": "INLINE",
      "value": "Review the code changes and suggest improvements"
    }
  ],
  "model": "sonnet",
  "claude_args": "--max-turns 5",
  "integration": {
    "hash_id": "my_anthropic_integration"
  },
  "dangerously_skip_permissions": true
}'
STATUS
200 OK