AWS CodePipeline
Browse all Amazonaws APIs★ Only Publicly Available OpenAPI DocumentDeveloper ToolsCi Cdhmac39 EndpointsREST
For Agents
Create and operate AWS CodePipeline release pipelines, start executions, retry failed stages, and approve gated transitions. Backed by 39 endpoints covering pipelines, executions, action types, and webhooks.
Use for: I need to create a release pipeline with source, build, and deploy stages, Start a CodePipeline execution for the latest source revision, Retry a failed stage in a CodePipeline execution, Approve a manual approval action in a pipeline
Not supported: Does not handle source compilation, container builds, or deployment execution itself — use for pipeline definition, execution, and stage approval orchestration only.
Jentic publishes the only available OpenAPI specification for AWS CodePipeline, keeping it validated and agent-ready. CodePipeline orchestrates multi-stage release pipelines with source, build, test, deploy, and approval actions, integrating with CodeCommit, GitHub, S3, ECR, CodeBuild, CodeDeploy, ECS, Lambda, and CloudFormation. The API covers pipelines, stages, executions, action types, webhooks, and manual approvals, so an agent can create pipelines, start executions, retry failed stages, and approve gated transitions.
Install Jentic One Beta
Connect the AWS CodePipeline to your agent
Jentic One is a self-hosted execution layer for AI agents. It lets your agent call the AWS CodePipeline, or any other public or private API you need. You set the rules, the agent never sees your credentials, and every call is logged.
Two steps, two machines. Install the instance in a safe environment, then register your agent from wherever it runs.
1
Step 1: Jentic One Host machine
# On the machine that will host your Jentic One instance:
curl -fsSL https://raw.githubusercontent.com/jentic/jentic-one/main/tools/install.sh | sh2
Step 2: Agent machine
# On the machine where your agent runs (keep this separate from the instance):
curl -fsSL https://raw.githubusercontent.com/jentic/jentic-one/main/tools/install.sh | sh
jentic register # connects your agent to your Jentic One instanceJentic One is in public beta. The setup above keeps your agent separate from the instance, which is what you want before using real credentials: an agent running as the same OS user as Jentic One can read its stored keys directly. Just evaluating? A single local install is fine to start. See the secure deployment guide for the tiers.
Capabilities
What an agent can do with AWS CodePipeline API.
Create and update CodePipeline pipelines with source, build, test, deploy, and approval actions
Start, list, and inspect pipeline executions and per-action execution details
Retry failed stages with RetryStageExecution and approve manual approval actions
Enable and disable stage transitions to pause or resume pipeline flow
Register and deregister webhooks that trigger pipelines from external sources
Define custom action types and put job results from custom action workers
Tag pipelines for cost allocation and ownership
Use Cases
Patterns agents use AWS CodePipeline API for, with concrete tasks.
★ End-to-End Release Pipeline Creation
Platform teams use CreatePipeline to define a YAML-style structure of stages and actions that pulls from CodeCommit or GitHub, builds with CodeBuild, runs tests, and deploys via CodeDeploy or CloudFormation. The single API call provisions the full pipeline. UpdatePipeline applies subsequent edits, and GetPipeline returns the current definition for diffing against version control.
Create a pipeline named web-release with three stages: Source from CodeCommit, Build with CodeBuild project web-build, and Deploy with CodeDeploy application web-svc
Manual Approval Gate Automation
Many production pipelines pause at a manual approval action so a human approves before deployment. PutApprovalResult lets an agent submit the approval (or rejection) along with a comment once external checks pass — for example, a Jira ticket transition, a successful canary, or a security scan result. ListPipelineExecutions and GetPipelineState reveal which approvals are pending so the agent knows where to act.
Approve the manual approval action ProdGate in stage Production of pipeline web-release with comment 'Canary metrics OK'
Failure Recovery and Stage Retry
When a stage fails on a transient infrastructure issue, RetryStageExecution restarts only the failed actions in that stage without re-running upstream stages. ListActionExecutions returns per-action status with errorDetails so an agent can decide whether retry, skip, or rollback is appropriate. DisableStageTransition pauses the pipeline at a known-good point during incidents.
Retry the failed stage Deploy in pipeline web-release for execution exec-abc with retryMode FAILED_ACTIONS
AI Agent Integration via Jentic
Through Jentic, an AI agent uses CodePipeline by searching for an intent like 'start a CodePipeline execution', loading the StartPipelineExecution schema, and executing it with a pipeline name. Jentic signs the SigV4 request server-side using IAM credentials in the MAXsystem vault.
Search Jentic for 'start a CodePipeline execution', load the StartPipelineExecution schema, and execute it for pipeline web-release
Key Endpoints
39 endpoints — jentic publishes the only available openapi specification for aws codepipeline, keeping it validated and agent-ready.
METHOD
PATH
DESCRIPTION
POST
/#X-Amz-Target=CodePipeline_20150709.CreatePipeline
Create a release pipeline
POST
/#X-Amz-Target=CodePipeline_20150709.GetPipeline
Get a pipeline definition
POST
/#X-Amz-Target=CodePipeline_20150709.GetPipelineState
Get the current state of every stage
POST
/#X-Amz-Target=CodePipeline_20150709.ListPipelineExecutions
List recent executions for a pipeline
POST
/#X-Amz-Target=CodePipeline_20150709.GetPipelineExecution
Get details of a single execution
POST
/#X-Amz-Target=CodePipeline_20150709.DisableStageTransition
Pause a stage transition
POST
/#X-Amz-Target=CodePipeline_20150709.EnableStageTransition
Resume a paused stage transition
POST
/#X-Amz-Target=CodePipeline_20150709.ListPipelines
List pipelines in the account
POST
/#X-Amz-Target=CodePipeline_20150709.CreatePipeline
Create a release pipeline
POST
/#X-Amz-Target=CodePipeline_20150709.GetPipeline
Get a pipeline definition
POST
/#X-Amz-Target=CodePipeline_20150709.GetPipelineState
Get the current state of every stage
POST
/#X-Amz-Target=CodePipeline_20150709.ListPipelineExecutions
List recent executions for a pipeline
POST
/#X-Amz-Target=CodePipeline_20150709.GetPipelineExecution
Get details of a single execution
POST
/#X-Amz-Target=CodePipeline_20150709.DisableStageTransition
Pause a stage transition
POST
/#X-Amz-Target=CodePipeline_20150709.EnableStageTransition
Resume a paused stage transition
POST
/#X-Amz-Target=CodePipeline_20150709.ListPipelines
List pipelines in the account
Why Jentic?
Three things that make agents converge on Jentic-routed access.
Credential isolation
AWS access key ID and secret are stored encrypted in the Jentic MAXsystem vault. Jentic computes the SigV4 signature server-side for every CodePipeline call, so raw IAM credentials never enter the agent's context.
Intent-based discovery
Agents search by intent (e.g. 'start a CodePipeline execution' or 'approve a pipeline action') and Jentic returns the matching CodePipeline operation with its input schema, so the agent calls StartPipelineExecution or PutApprovalResult directly without browsing AWS docs.
Time to first call
Direct AWS SDK integration: 2-4 days for SigV4, IAM, pipeline JSON authoring, and execution polling. Through Jentic: under 1 hour — search, load schema, execute.
Related APIs
Alternatives and complements available in the Jentic catalogue.
FAQs
Specific to using AWS CodePipeline API through Jentic.
Why is there no official OpenAPI spec for AWS CodePipeline?
AWS does not publish an OpenAPI specification. Jentic generates and maintains this spec so that AI agents and developers can call AWS CodePipeline via structured tooling. It is validated against the live API and kept up to date. Get started at https://app.jentic.com/sign-up.
What authentication does the AWS CodePipeline API use?
CodePipeline uses AWS Signature Version 4 (HMAC) signed requests with an IAM access key and secret. Through Jentic, IAM credentials live encrypted in the MAXsystem vault and the SigV4 signature is computed server-side, so the agent context never holds the raw secret.
Can I approve a manual approval action through the API?
Yes. Call PutApprovalResult with pipelineName, stageName, actionName, the token from GetPipelineState, and a result object containing status (Approved or Rejected) and a summary. The pipeline immediately resumes or terminates based on the status.
How do I retry a failed stage in a CodePipeline execution?
Call RetryStageExecution with the pipelineName, stageName, pipelineExecutionId, and retryMode FAILED_ACTIONS. CodePipeline restarts only the failed actions in that stage and continues forward when they succeed.
What are the rate limits for the AWS CodePipeline API?
AWS enforces per-account, per-region throttling on CodePipeline; ListPipelineExecutions and similar list calls are rate-limited per account. Throttled requests return ThrottlingException — the AWS SDKs that Jentic wraps implement exponential backoff automatically.
How do I start a CodePipeline release through Jentic with an AI agent?
Run pip install jentic, then have the agent search for 'start a CodePipeline execution', load the StartPipelineExecution schema, and execute it with the pipeline name. Poll GetPipelineState for stage status and use PutApprovalResult to clear manual approval gates as criteria are met.
GET STARTED