High Risk →

run_rollback_pipeline

Run a rollback pipeline for a CircleCI project. This tool guides you through the full rollback process, adapting to the information you provide and prompting for any missing details. **Initial Requirements:** - You need either a `projectSlug` (from `listFollowedProjects`) or a `proj...

High parameter count (10 properties)

Part of the CircleCI MCP Server MCP server. Enforce policies on this tool with Intercept, the open-source MCP proxy.

AI agents invoke run_rollback_pipeline to trigger processes or run actions in CircleCI MCP Server. Execute operations can have side effects beyond the immediate call -- triggering builds, sending notifications, or starting workflows. Rate limits and argument validation are essential to prevent runaway execution.

run_rollback_pipeline can trigger processes with real-world consequences. An uncontrolled agent might start dozens of builds, send mass notifications, or kick off expensive compute jobs. Intercept enforces rate limits and validates arguments to keep execution within safe bounds.

Execute tools trigger processes. Rate-limit and validate arguments to prevent unintended side effects.

circleci-public-mcp-server-circleci.yaml
tools:
  run_rollback_pipeline:
    rules:
      - action: allow
        rate_limit:
          max: 10
          window: 60
        validate:
          required_args: true

See the full CircleCI MCP Server policy for all 16 tools.

Tool Name run_rollback_pipeline
Category Execute
Risk Level High

View all 16 tools →

Agents calling execute-class tools like run_rollback_pipeline have been implicated in these attack patterns. Read the full case and prevention policy for each:

Browse the full MCP Attack Database →

Other tools in the Execute risk category across the catalogue. The same policy patterns (rate-limit, validate) apply to each.

run_rollback_pipeline is one of the high-risk operations in CircleCI MCP Server. For the full severity-focused view — only the high-risk tools with their recommended policies — see the breakdown for this server, or browse all high-risk tools across every MCP server.

What does the run_rollback_pipeline tool do? +

Run a rollback pipeline for a CircleCI project. This tool guides you through the full rollback process, adapting to the information you provide and prompting for any missing details. **Initial Requirements:** - You need either a `projectSlug` (from `listFollowedProjects`) or a `projectID`. The tool will automatically resolve the project information from either of these. **Typical Flow:** 1. **Start:** User initiates a rollback request. 2. **Project Selection:** If project id or project slug are not provided, call `listFollowedProjects` to get the list of projects the user follows and present the full list of projects to the user so that they can select the project they want to rollback. 3. **Project Information:** Provide either `projectSlug` or `projectID`. The tool will automatically resolve the project information as needed. 4. **Version Selection:** If component environment and version are not provided, call `listComponentVersions` to get the list of versions for the selected component and environment. If there is only one version, proceed automatically and do not ask the user to select a version. Otherwise, present the user with the full list of versions and ask them to select one. Always return all available values without categorizing them. 5. **Rollback Reason** ask the user for an optional reason for the rollback (e.g., "Critical bug fix"). Skip this step is the user explicitly requests a rollback by workflow rerun. 6. **Rollback pipeline check** if the tool reports that no rollback pipeline is defined, ask the user if they want to trigger a rollback by workflow rerun or suggest to setup a rollback pipeline following the documentation at https://circleci.com/docs/deploy/rollback-a-project-using-the-rollback-pipeline/. 7. **Confirmation:** Summarize the rollback request and confirm with the user before submitting. 8. **Pipeline Rollback:** if the user requested a rollback by pipeline, call `runRollbackPipeline` passing all parameters including the namespace associated with the version to the tool. 9. **Workflow Rerun** If the user requested a rollback by workflow rerun, call `rerunWorkflow` passing the workflow ID of the selected version to the tool. 10.**Completion:** Report the outcome of the operation. **Parameters:** - `projectSlug` (optional): The project slug from `listFollowedProjects` (e.g., "gh/organization/project"). Either this or `projectID` must be provided. - `projectID` (optional): The CircleCI project ID (UUID). Either this or `projectSlug` must be provided. - `environmentName` (required): The target environment (e.g., "production", "staging"). - `componentName` (required): The component to rollback (e.g., "frontend", "backend"). - `currentVersion` (required): The currently deployed version. - `targetVersion` (required): The version to rollback to. - `namespace` (required): The namespace of the component. - `reason` (optional): Reason for the rollback. - `parameters` (optional): Additional rollback parameters as key-value pairs. **Behavior:** - If there are more than 20 environments or components, ask the user to refine their selection. - Never attempt to guess or construct project slugs or URLs; always use values provided by the user or from `listFollowedProjects`. - Do not prompt for missing parameters until versions have been listed. - Do not call this tool with incomplete parameters. - If the selected project lacks rollback pipeline configuration, provide a definitive error message without suggesting alternative projects. **Returns:** - On success: The rollback ID or a confirmation in case of workflow rerun. - On error: A clear message describing what is missing or what went wrong. - If the selected project does not have a rollback pipeline configured: The tool will provide a clear error message specific to that project and will NOT suggest trying another project. **Important Note:** - This tool is designed to work only with the specific project provided by the user. - If a project does not have rollback capability configured, the tool will NOT recommend trying other projects. - The assistant should NOT suggest trying different projects when a project lacks rollback configuration. - Each project must have its own rollback pipeline configuration to be eligible for rollback operations. - When a project cannot be rolled back, provide only the configuration guidance for THAT specific project. - The tool automatically resolves project information from either `projectSlug` or `projectID`. If no version is found, the tool will suggest the user to set up deploy markers following the documentation at: https://circleci.com/docs/deploy/configure-deploy-markers/ . It is categorised as a Execute tool in the CircleCI MCP Server MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.

How do I enforce a policy on run_rollback_pipeline? +

Add a rule in your Intercept YAML policy under the tools section for run_rollback_pipeline. You can allow, deny, rate-limit, or validate arguments. Then run Intercept as a proxy in front of the CircleCI MCP Server MCP server.

What risk level is run_rollback_pipeline? +

run_rollback_pipeline is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.

Can I rate-limit run_rollback_pipeline? +

Yes. Add a rate_limit block to the run_rollback_pipeline rule in your Intercept policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.

How do I block run_rollback_pipeline completely? +

Set action: deny in the Intercept policy for run_rollback_pipeline. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.

What MCP server provides run_rollback_pipeline? +

run_rollback_pipeline is provided by the CircleCI MCP Server MCP server (CircleCI-Public/mcp-server-circleci). Intercept sits as a proxy in front of this server to enforce policies before tool calls reach the server.

Let agents act without letting them run wild.

Deterministic policy on every MCP tool call. Per-identity grants. Full audit log.

// GET IN TOUCH

Have a question or want to learn more? Send us a message.

Message sent.

We'll get back to you soon.