YAML for Codex CLI

View as Markdown

Run OpenAI Codex CLI AI assistant.

YAML parameters for Codex CLI

PARAMETERS

promptsrequired(string | object)[]

List of prompts for the agent. Each entry is either an inline string or a file reference (type INLINE or FILE).

typerequiredstring

Value: CODEX_CLI

actionrequiredstring

Unique identifier for the action within the pipeline.

sandbox_modestring enum

Sandbox mode: READ_ONLY, WORKSPACE_WRITE, or FULL_ACCESS

Allowed enum:

READ_ONLY,

WORKSPACE_WRITE,

FULL_ACCESS

integrationstring

Integration identifier for authenticating with the AI provider.

modelstring

The Codex model to use (e.g. codex-mini).

skillsstring[]

List of agent skills to install before the run. Each entry is an owner/repo, a repository URL, or a local path.

mcp_serversRecord<string, McpServerYaml>

MCP servers to make available to the agent, keyed by server name. Each server is either http/sse (url, headers) or stdio (command, args, env).

working_directorystring

The directory in which the pipeline filesystem will be mounted.

commandsstring

The commands that will be executed.

cached_dirsstring[]

The dependencies and directories to be cached and available to every execution in this pipeline.

mount_filesystem_disableboolean

If set to true, the filesystem will not be mounted in the container.

volume_mappingsstring[]

The path preceding the colon is the filesystem path (the folder from the filesystem to be mounted in the container). The path after the colon is the container path (the path in the container, where this filesystem will be located).

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.

run_as_userstring

All build commands are run as the default user defined in the selected Docker image. Can be set to another username (on the condition that this user exists in the selected image).

entrypointstring

The entrypoint to use for the Docker container.

reset_entrypointboolean

If set to true, resets the default entrypoint set by the image.

cache_base_imageboolean

If set to true, the cached version of the image is used, instead of being pulled each time.

ignore_image_pull_failuresboolean

If set to true, use cached image on timeouts (only for official images).

export_container_pathstring

The path in the container to export.

docker_image_namestring

The name of the Docker image.

docker_image_tagstring

The tag of the Docker image.

image_locationstring enum

The location of the image used by the action.

Allowed enum:

PUBLIC_REGISTRY,

PRIVATE_REGISTRY,

ACTION,

ARTIFACT_REGISTRY

docker_registrystring enum

The type of registry from which the image is retrieved.

Allowed enum:

NONE,

DOCKER_HUB,

AMAZON_ECR,

GOOGLE_GCR,

GOOGLE_ARTIFACT_REGISTRY,

OTHER,

GIT_HUB_CONTAINER_REGISTRY,

ARTIFACT_REGISTRY,

DIGITAL_OCEAN_CONTAINER_REGISTRY

artifactstring

The identifier of the artifact from artifact registry.

use_image_from_actionboolean

If set to true, the image from another action will be used.

docker_build_action_namestring

The name of the action from which the Docker image will be used.

registrystring

The URL of the Docker registry.

insecure_registryboolean

If set to true, the registry will be accessed over HTTP instead of HTTPS.

target_stagestring

The target stage of the Dockerfile.

do_not_prune_imagesboolean

If set to true, images will not be pruned after the build.

loginstring

Username for a private registry. Used only when docker_registry is OTHER; for managed registries use integration instead.

passwordstring

Password for a private registry. Used only when docker_registry is OTHER; for managed registries use integration instead.

servicesServiceYaml[]

The list of services attached to the build environment.

visual_tests_suitestring

The name of the visual test suite to run tests against.

targets(string | TargetYaml)[]

The list of target servers or clusters.

trigger_timestring enum

Specifies when the action should be executed.

Allowed enum:

ON_EVERY_EXECUTION,

ON_FAILURE,

ON_BACK_TO_SUCCESS,

ON_WARNING,

ON_WAIT_FOR_APPROVE,

ON_TERMINATE

loopstring[]

The list of variables for dynamic action execution. The action runs once for each value.

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.

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_intervalinteger

Delay time between auto retries in seconds.

retry_countinteger

Number of retries if the action fails.

Constraints: ≤ 10

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

trigger_conditionsTriggerConditionYaml[]

The list of trigger conditions to meet so that the action can be triggered.

variablesVariableYaml[]

The list of variables you can use in the action.

Last modified on Sep 17, 2026

CodeDeploy Composer Publish

YAML examples for Codex CLI

Run Codex CLI with all options

yaml
- action: "Run Codex CLI"
  type: "CODEX_CLI"
  trigger_time: "ON_EVERY_EXECUTION"
  integration: "my-openai-integration"
  prompts:
    - "Review the code and suggest improvements"
    - file: ".buddy/codex-prompt.md"
  model: "codex-mini"
  skills:
    - buddy-works/security-audit
  mcp_servers:
    github:
      type: "http"
      url: "https://api.githubcopilot.com/mcp/"
      headers:
        Authorization: "Bearer $GITHUB_TOKEN"
    filesystem:
      type: "stdio"
      command: "npx"
      args:
        - "-y"
        - "@modelcontextprotocol/server-filesystem"
        - "/buddy/my-repo"
  sandbox_mode: "WORKSPACE_WRITE"

Minimal Codex CLI

yaml
- action: "Codex Review"
  type: "CODEX_CLI"
  trigger_time: "ON_EVERY_EXECUTION"
  integration: "my-openai-integration"
  prompts:
    - "Review the code"