Integrations
manent sits between your CI, your artifact repositories and your targets, and talks to the tools around them: where people discuss releases, where changes are approved, where secrets and configuration live, where metrics decide whether a release is healthy. Each guide below is a step-by-step setup for one tool: what to prepare on its side, what to write in manent.yml, and how to check that it works.
One rule for every integration
Credentials are always secret references (vault:…, file:…, env:…), read when they are used and never written to logs, outputs or the deploy history. Deliveries to external systems are durable steps: a deploy resumed after a crash neither loses nor duplicates them, and a failed notification never fails a deploy.
Chat
| Tool |
What you get |
Guide |
| Slack |
Deploy started and outcome in a channel, one message updated in place; approval requests with Approve and Reject buttons that respect manent permissions. |
Slack |
| Microsoft Teams |
Adaptive Cards for deploys and approvals, with a button that opens the deploy in the web interface. |
Microsoft Teams |
Change management
| Tool |
What you get |
Guide |
| ServiceNow |
A change request per deploy, opened when it starts and closed with its outcome; optionally the deploy waits for the change to be approved. |
ServiceNow |
| Jira |
The same with a Jira issue (Cloud or Data Center), with transitions and an outcome comment. |
Jira |
CI/CD and infrastructure as code
| Tool |
What you get |
Guide |
| GitHub Actions |
A composite action that deploys, validates descriptors or candidate configurations, with outputs and job summary. |
GitHub Actions |
| GitLab CI |
A template included with typed inputs and a dotenv report for the following jobs. |
GitLab CI |
| Terraform |
The release in service as a resource, drift in the plan, and configuration checks that fail the plan. |
Terraform |
| Any CI |
manent deploy --outputs/--summary, exit codes and idempotent --id. |
Deploy from a CI pipeline |
Configuration, secrets and artifacts
| Tool |
What you get |
Guide |
| HashiCorp Consul |
The environment configuration in the KV store, read at every deploy with its revision. |
HashiCorp Consul |
| Git |
The environment configuration in a Git repository, read at a commit whose SHA is recorded on the deploy; optional signed commits. |
Git repository |
| HashiCorp Vault |
Secrets from KV v1 or v2, with token, AppRole or Kubernetes authentication. |
HashiCorp Vault |
| Nexus, Artifactory, OCI registries |
Allowed artifact sources with credentials per URL prefix and private CAs. |
Artifact repositories |
Identity
| Tool |
What you get |
Guide |
| Keycloak, Microsoft Entra ID, Okta (any OIDC provider) |
User sign-in for the API, the CLI and the web interface, with groups mapped to roles. |
OIDC identity providers |
Metrics and health gates
| Tool |
What you get |
Guide |
| Prometheus, Alertmanager |
Health gates on PromQL queries and active alerts; monitoring of manent itself. |
Prometheus and Alertmanager |
| Datadog, New Relic, Elastic, Amazon CloudWatch, Azure Monitor |
The same health gates and canary analysis on the metrics system you already use. |
Metrics systems |
| OpenTelemetry |
A trace per deploy, from the CI request to every step on the worker. |
Observability |
Logs and audit
| Tool |
What you get |
Guide |
| Splunk, Elastic, QRadar, any syslog collector, journald |
Every log record and audit entry as RFC 5424 with structured data, or as journal fields. |
SIEM and syslog |
Load balancers and traffic routers
These are configured per environment, in the environment configuration (traffic), and described in the guides of the strategies that use them:
| Tool |
Used for |
Guide |
| HAProxy, F5 BIG-IP, NGINX Plus, NGINX, AWS ELB, IIS ARR, Windows NLB |
Draining batches of targets during a wave rollout. |
Load balancer draining |
| Gateway API, Istio, AWS ALB, HAProxy, F5, NGINX Plus, Lambda aliases, Cloud Run, App Service, Container Apps |
Moving a percentage of traffic to the candidate version. |
Percentage rollouts |
Anything else
- Webhooks. Callbacks send an HTTP event, optionally signed with HMAC, for every step and for the outcome of every deploy: the generic way to feed a dashboard, a bot or an internal portal.
- HTTP APIs in the pipeline. A custom step of type
http calls an external system during the deploy, for example to purge a CDN, and can wait for its outcome.
- Your own platform. An external plugin releases to targets manent does not support, in any language that speaks gRPC.
The configuration keys of every integration are listed in the reference: Chat and change management, CI and infrastructure as code, Configuring manent.