ceven

Seqera

Launches and monitors Nextflow pipelines on demand, tracks compute resource usage across clouds, and alerts you when genomic analysis runs fail or complete.

Try Seqera in Ceven

Ask Ceven anything
Standard

Why use Ceven?

  1. AI native Seqera integration

    • Describe the outcome and Ceven picks the right Seqera calls, fills the parameters, and checks the result.
    • Structured, agent friendly tool schemas so each call runs reliably instead of by guesswork.
    • Rich coverage for reading, writing, and querying your Seqera data, across all 14 of its actions.

  2. Managed auth

    • Built in OAuth with automatic token refresh and rotation.
    • One place to manage, scope, and revoke Seqera access.
    • Per user and per environment credentials instead of shared keys.

  3. Agent optimized design

    • Actions are tuned from real success and error rates so reliability climbs over time.
    • Full execution logs so you always know what ran in Seqera, when, and on whose behalf.
    • The agent pauses and asks when Seqera is unclear instead of plowing ahead.

  4. Enterprise grade security

    • Fine grained access so you control which agents and people can reach Seqera.
    • Least privilege by default, read scopes first and only the writes a workflow needs.
    • A full audit trail of every Seqera action to support review and sign off.

Supported tools

Every action Ceven's agents can run on Seqera, and when to use it.

Launch pipeline
Use this to start a new Nextflow run. Specify the pipeline ID, input parameters, and the compute profile to use.
Get run status
Pull the current state of a specific pipeline run, including whether it is running, completed, or failed.
List runs
Pull a list of all recent pipeline executions, optionally filtered by user, date, or pipeline name.
Get run logs
Retrieve the execution logs for a specific run to diagnose errors or verify process completion.
Stop run
Terminate a running pipeline immediately to save compute costs or stop a misconfigured run.
List pipelines
Pull all available pipeline definitions and versions accessible to the connected account.
Get compute profile
Pull details on a specific compute configuration, including instance types and cloud provider settings.
Update run tags
Add or modify metadata tags on a pipeline run for better organization and billing tracking.
Search runs
Query pipeline runs using specific metadata, tags, or parameter values.
Get resource usage
Pull CPU and memory consumption metrics for a specific run to identify bottlenecks.
List users
Pull a list of users with access to the Seqera Platform environment.
Get pipeline version
Pull the specific git commit or version tag for a pipeline definition.

12 actions · scroll to see them all

Frequently asked questions

Ceven monitors the run status via the Seqera API. When a run transitions to a failed state, the agent can be configured to immediately pull the logs for the failing process. It then parses these logs for known error strings, such as out of memory errors or missing input files. Based on the findings, Ceven can either notify a human operator with the specific error snippet or attempt a predefined recovery action, such as relaunching the pipeline with a larger compute profile. This prevents runs from sitting in a failed state unnoticed for hours, ensuring that critical data analysis timelines are met without constant manual polling of the Seqera dashboard.
Ceven can help control costs by monitoring the resource usage and duration of your runs. You can build a workflow that lists all currently running pipelines and flags any that have exceeded a specific time limit or budget. Because Ceven has the ability to stop runs, it can automatically terminate zombie processes or runs that are clearly looping. By pulling the compute profile details, the agent can also suggest moving certain workloads to spot instances if the pipeline is fault tolerant. This gives you a programmatic layer of governance over your cloud spend that goes beyond the basic alerts provided within the Seqera Platform interface.
The account connecting Ceven to Seqera must have the necessary Role Based Access Control permissions to perform the requested actions. If the user only has read access in Seqera, the agent will be unable to launch or stop pipelines, and those actions will return an authorization error. We recommend using a dedicated service account with a custom role that is scoped specifically to the pipelines and compute profiles the agent needs to manage. This ensures the principle of least privilege is maintained while allowing the agent to automate the operational side of your bioinformatics workflows without granting full administrative rights to the entire platform.
No. Ceven interacts with the Seqera Platform API to manage the orchestration of pipelines, but it does not ingest or store the actual biological data files. The agent sees metadata, such as file names, run IDs, and execution logs, but the heavy data remains in your secure storage buckets or on your local compute cluster. Ceven acts as the control plane that tells Seqera what to do and asks it how it is going. Your data privacy and compliance posture remain unchanged because the data flow happens directly between your storage and the Nextflow executors, never passing through the Ceven infrastructure.
Ceven implements an exponential backoff strategy to handle Seqera API rate limits. If the platform returns a 429 too many requests error, the agent pauses and retries the call with increasing delays. For high volume environments with thousands of concurrent runs, we recommend using the list runs action with tight filters to reduce the payload size and frequency of calls. Note that some Seqera tiers have stricter API quotas than others, which may result in slower update intervals for your workflow dashboards during peak usage. The agent will log these throttling events so you can decide if a tier upgrade is necessary for your scale.
Ceven can interact with the pipeline versions registered in Seqera, but it does not perform git operations directly on your source code. You use Ceven to specify which version of a pipeline should be launched for a particular run. For example, you can create a workflow that launches version 1.0 for production samples and version 1.1 for a small test batch to compare results. The agent can then pull the logs and resource metrics for both versions and present a side by side comparison. This allows you to manage the rollout of new pipeline versions through the Seqera API without leaving your orchestration layer.
When a run is interrupted, such as when a spot instance is reclaimed, Seqera updates the run status. Ceven detects this state change through its monitoring loop. Depending on your workflow logic, the agent can trigger a restart of the pipeline from the last successful checkpoint, which is a native feature of Nextflow. The agent handles the API call to resume the run and then monitors it to ensure it progresses past the point of failure. This automation removes the need for a human to manually monitor cloud interruptions and manually hit the resume button in the Seqera UI.
There is no hard limit on the number of pipelines Ceven can monitor, but performance depends on the polling frequency you set in your workflow. Because Ceven relies on the Seqera API, monitoring thousands of runs in real time can lead to API throttling. To optimize this, we recommend using a search runs action to only track active or failing runs rather than listing every single run in your history. By narrowing the scope of the read calls, the agent can maintain a high level of responsiveness without hitting the platform limits, ensuring that critical alerts are delivered the moment a pipeline fails.

Alternatives to Seqera

Other tools that solve a similar problem. Ceven supports these too, so you can switch or run more than one at once.

DNAnexus logoDNAnexusTerra logoTerraSeven Bridges logoSeven Bridges

Try Ceven on your stack

Plug Ceven on top of the tools you already run. Connect Seqera and the rest of your stack, describe the outcome, and its agents handle the work end to end, days of it in minutes.

Get started for free