Connectors · Developer & infrastructure
Harness Platform MCP server for AI agents.
actionlake is an MCP gateway: one remote MCP server that connects Claude and other MCP clients to Harness Platform and the rest of your systems. It gives your agent 2,994 Harness Platform actions (1,545 reads, 1,092 writes and 357 that wait for a confirmation) and holds the Harness Platform credential itself, so the agent never does. It paces Harness Platform's limits and records every call.
100,000 free calls a month. No card, no email, no password: press Start a free sandbox on the next screen and keep the key it shows you.
claude mcp add --transport http actionlake https://connect.actionlake.com/mcpdo { action: "harness.account_roles.list_roles_acc", intent: "check which Harness Platform account is connected"}# List Roles · read · 1 upstream call- 2,994actionsin one signed file
- 1,545readlook, never change
- 1,092writecreate and update
- 357wait for a yesdestructive, confirmed first
Set up
Connect Harness Platform in three steps.
- 01
Add actionlake to Claude.
Run the command once in Claude Code; its consent page can create a free sandbox: 100,000 calls a month, no card. In the Claude app, add a custom connector with the URL below and paste your key on the consent page. Creating a new sandbox from the Claude app is not yet tested by us.
https://connect.actionlake.com/mcp - 02
Connect Harness Platform.
API key. No consent screen: your agent can connect it for you, but the key then passes through your chat. To keep it out of the model's context, paste it in the console instead. actionlake stores what you paste write-only, sealed under your tenant's key.
api_keyHarness API key token: in Harness open your profile (or Account Settings > Access Control > Service Accounts),…account_idYour Harness Platform account idhostWhere your Harness Platform lives: your host
account { op: "connections.requirements", service: "harness" } # free: exactly what to paste, and what is found for you - 03
Make the first call. Check before a write.
Ask in plain words; Claude picks the action. Before anything changes,
checkvalidates the call and prices it, for free.do { action: "harness.account_roles.list_roles_acc" }# list roles · read · 1 upstream callcheck { action: "harness.account_roles.create_role_acc", input: { … } }# free: inputs, permissions, cost and confirmations, nothing sent
Harness Platform at a glance
- Sign-in
- API key. No consent screen: your agent can connect it for you, but the key then passes through your chat. To keep it out of the model's context, paste it in the console instead.
- You provide
api_key- Setup fields
account_idhost- Limits we pace to
- 300 calls a minute, per connection
- API covered
- Harness Platform: CI/CD pipelines, GitOps, Code, Artifact Registry, Feature Flags, Chaos, Cloud Cost, IaCM, STO, SRM and platform administration API Harness APIs (OpenAPI 3.0.3, spec 1.0)
app.harness.io - Built from
- Harness Platform's documentation, read 23 October 2026apidocs.harness.ioapidocs.harness.io
- Service file
- Version 1.0.0, signed (Ed25519)
- Cached reads
- 1,288 of 1,545 reads, for up to 5 minutes; a write clears what it affects
- Shared entities
- 3, so ask can line Harness Platform up with your other systems
- Data
- Processed and stored in Sydney, Australia. No US region yet
Capabilities
What an AI agent can do in Harness Platform.
2,994 actions in 485 areas. Every one carries its effect, so a key can allow the reads and hold the rest.
- 1,545readlook, never change
- 1,092writecreate and update
- 357destructivewait for a confirmation
Ansible
7839 read · 30 write · 9 destructive
List Ansible Activities · List Ansible Activities on account level · List Ansible Activities on org level · List data
Pullreq
5219 read · 24 write · 9 destructive
List pull requests · Get pull request · Create pull request · Update pull request
Repository
4722 read · 20 write · 5 destructive
List repositories · Get repository · Create repository · Update repository
Workspaces
4220 read · 17 write · 5 destructive
List all workspace tags · List workspaces associated with a template ID · List data · List modules
Monitored services
3726 read · 9 write · 2 destructive
fetch reconciliation status for template referenced monitored services · Get notification rules for MonitoredService · Cvget anomalies summary · Get all monitored services with health sources
Applications
3520 read · 11 write · 4 destructive
List resource actions · Get ManagedResources · List LegacyApplicationSets · List attributes for applications
Module registry
3515 read · 16 write · 4 destructive
List connectors · List module example tree · List executions for a specific module · List module providers
Frontend
3428 read · 6 write · 0 destructive
Returns daily trend data of issue counts by status for a given time interval · Returns issue summaries that are going to expire · Provides baseline execution data needed by the Overview page · Provides historical issue data needed by the Overview page
112 more areas · 2,634 actions
- Configuration variable sets 33
- Chaos infrastructure 31
- Experiments 31
- Repositories 24
- Exemptions 22
- Provider registry 22
- Artifacts 21
- Executions 21
- Labels 21
- Remediation 21
- Variable sets 21
- Budgets 20
- Filter 20
- Templates 20
- User 20
- User group 20
- Authentication settings 19
- Ldap settings 19
- Notification rules 19
- Orchestration executions 19
- Notification channels 18
- Pipeline execution details 18
- Services 18
- Test targets 18
- Agents 17
- Clusters 17
- Entities 17
- Registry V3 versions 17
- Connectors 16
- Webhook 16
- Environments 15
- Onboarding 15
- Pipeline 15
- Recommendations 15
- Rule 15
- Pipeline input set 14
- Registry V2 14
- Registry V3 firewall 14
- Spaces 14
- Triggers 14
- Auto stopping rules 13
- Dashboards 13
- File store 13
- Rule executions 13
- Pipeline step 12
- Pipelines 12
- Recommendation jira 12
- Registry V3 lifecycle 12
- Releases 12
- Secrets 12
- Settings 12
- Usage 12
- Variables 12
- Backstage env variable 11
- Commitment orchestrator setup ap is 11
- Fault templates 11
- Orchestration processes 11
- Overrides 11
- Probes 11
- Project mappings 11
- Risks 11
- Account template 10
- Application set 10
- Experiment templates 10
- Freeze crud 10
- Org template 10
- Organization 10
- Project template 10
- Registry V3 packages 10
- Role assignments 10
- Rules 10
- Scans 10
- Downtime 9
- External data provider 9
- Infrastructure 9
- Input sets 9
- Integration 9
- Orchestration activities 9
- Pipeline execute 9
- Registry V3 registries 9
- Rule enforcement 9
- SL OS dashboard 9
- System 9
- Action templates 8
- Actions 8
- Budget groups 8
- Catalog custom properties 8
- Chaos recommendations 8
- Cost categories 8
- Data job status 8
- Delegate setup resource 8
- Faults 8
- Feature flags 8
- Governance overview 8
- Infrastructures 8
- Integrations 8
- Ip allowlist 8
- Perspectives 8
- Plugin info 8
- Policies 8
- Registry V3 signing keys 8
- Release groups 8
- Replication 8
- Service account 8
- Streaming destinations 8
- Token 8
- Webhooks 8
- Account banner 7
- Account services 7
- Accounts 7
- Aggregation rules 7
- 366 smaller areas 1,196
Top actions
The Harness Platform API, shaped for agents.
The ids your agent types, what each does, and the one name it shares with other systems. 44 of Harness Platform's 44 action names are shared with at least one other connector, so an agent learns them once.
| Action | What it does | Effect | Same name in |
|---|---|---|---|
harness.account_roles.list_roles_accList Roles | List Roles. | read | identity.role.list184 connectors |
harness.repository.list_reposList repositories | List repositories. | read | devops.repository.list28 connectors |
harness.org_project.get_org_scoped_projectsList projects | List projects. | read | work.project.list109 connectors |
harness.account_roles.get_role_accRetrieve a Role | Retrieve a Role. | read | identity.role.get127 connectors |
harness.repository.get_repositoryGet repository | Get repository. | read | devops.repository.get26 connectors |
harness.user_group.get_user_group_listList the User Groups in an account/org/project | List the User Groups in an account/org/project. | read | identity.group.list262 connectors |
harness.account_roles.create_role_accCreate a Role | Create a Role. | write | identity.role.create105 connectors |
harness.user.get_current_user_infoGet Current User Info | Get Current User Info. | read | identity.current_user.get307 connectors |
harness.repository.create_repositoryCreate repository | Create repository. | write | devops.repository.create18 connectors |
harness.org_project.get_org_scoped_projectRetrieve a project | Retrieve a project. | read | work.project.get106 connectors |
harness.account_roles.update_role_accUpdate a Role | Update a Role. | write | identity.role.update98 connectors |
harness.account_roles.delete_role_accDelete a Role | Delete a Role. | destructive | identity.role.delete106 connectors |
Try it
Prompts to paste into Claude.
Each one names the Harness Platform actions it will call, so you know what it reads and what it changes before you send it.
- Readread
Give me a one-line status for every active Harness Platform project.
calls
harness.org_project.get_org_scoped_projects - Writewrite
Create a Role in Harness Platform. Show me exactly what you will send before you save it.
calls
harness.account_roles.create_role_acc
Governance for Harness Platform
Destructive actions wait for a yes.
357 Harness Platform actions are destructive. The first 10:
- Delete a Role
harness.account_roles.delete_role_acc - Soft delete repository
harness.repository.delete_repository - Delete a project
harness.org_project.delete_org_scoped_project - Delete an Environment
harness.project_environments.delete_environment - Delete a Feature Flag
harness.feature_flags.delete_feature_flag - Delete a Pipeline
harness.pipelines.delete_pipeline - Deletes a secret
harness.account_secret.delete_account_scoped_secret - Delete a Key
harness.keys.delete_key - Delete a Tag
harness.tags.delete_tag - Delete a rule
harness.rule.delete_rule
A destructive action runs only with confirm: true. Under the human policy the agent cannot confirm for itself: the run stops until a named person approves, in the console or from a signed link in an email. The cap on a key is refused before the call is sent, never after.
Delete a Role?
intent: the user asked me to delete a Role
account { op: "keys.create", name: "harness-agent", allowed_actions: ["harness.*"], cap_calls: 2000, confirm_policy: "human_for_writes", approval_rules: [ { match: "harness.account_roles.delete_role_acc", approver: "[email protected]" }, { match: "harness.repository.delete_repository", approver: "[email protected]" }, { match: "harness.org_project.delete_org_scoped_project", approver: "[email protected]" } ]}Cost
What Harness Platform costs on actionlake.
Every Harness Platform action is one upstream call, the only thing actionlake bills. Searching, check, cost, account and any call Harness Platform refused with 401, 403, 429 or a 5xx are free. Paid plans start at US$49 a month for 250,000 calls, with overage from US$0.30 down to US$0.18 per 1,000.
- 1upstream call per actionall 2,994 Harness Platform actions
- 2,994actions pageeach page of a long list is one call
- 1,288reads can be cacheda cache hit costs nothing
- 100,000calls a month, freepaid plans from US$49 a month for 250,000
Action index
Every Harness Platform action.
2,994 actions by area: the id your agent calls, what it does, and its effect.
Ansible78
harness.ansible.ansible_list_ansible_activitiesList Ansible Activitiesreadharness.ansible.ansible_list_ansible_activities_account_levelList Ansible Activities on account levelread- and 76 more in this area
Pullreq52
harness.pullreq.create_pull_reqCreate pull requestwriteharness.pullreq.get_pull_reqGet pull requestread- and 50 more in this area
Repository47
harness.repository.create_repositoryCreate repositorywriteharness.repository.get_repositoryGet repositoryread- and 45 more in this area
Workspaces42
harness.workspaces.workspaces_list_all_workspace_tagsList all workspace tagsreadharness.workspaces.workspaces_list_associated_workspacesList workspaces associated with a template IDread- and 40 more in this area
Monitored services37
harness.monitored_services.cvget_anomalies_summaryCvget anomalies summaryreadharness.monitored_services.get_all_monitored_services_with_health_sourcesGet all monitored services with health sourcesread- and 35 more in this area
Applications35
harness.applications.agent_application_service_list_resource_actionsList resource actionsreadharness.applications.agent_application_service_managed_resourcesGet ManagedResourcesread- and 33 more in this area
Module registry35
harness.module_registry.module_registry_list_connectorsList connectorsreadharness.module_registry.module_registry_list_module_example_treeList module example treeread- and 33 more in this area
Frontend34
harness.frontend.frontend_all_issues_trend_v2Returns daily trend data of issue counts by status for a given time intervalreadharness.frontend.frontend_expiring_exemptionsReturns issue summaries that are going to expireread- and 32 more in this area
Configuration variable sets33
harness.configuration_variable_sets.configuration_variable_sets_get_configuration_variable_set_attachments_account_levelGet Configuration Variable Set attachments on account levelreadharness.configuration_variable_sets.configuration_variable_sets_get_configuration_variable_set_attachments_org_levelGet Configuration Variable Set attachments on org levelread- and 31 more in this area
Chaos infrastructure31
harness.chaos_infrastructure.get_legacy_kubernetes_helm_infra_commandGet legacy kubernetes helm infra commandreadharness.chaos_infrastructure.get_v2_infrastructure_yamlPreview v2 infra Yamlread- and 29 more in this area
Experiments31
harness.experiments.get_chaos_experiment_runs_for_pipelineList the experiment runs for a pipeline executionreadharness.experiments.get_experiment_or_template_variableGet experiment or template inputsread- and 29 more in this area
Repositories24
harness.repositories.agent_repository_service_get_by_urlGet a repository configured in gitops by URL from Harnessreadharness.repositories.agent_repository_service_get_helm_chartsGetHelmCharts returns list of helm charts in the specified repositoryread- and 22 more in this area
Exemptions22
harness.exemptions.exemptions_export_issue_exemption_historyExport exemption history for issue as CSVreadharness.exemptions.exemptions_list_exemptionsList a collection of Exemptionsread- and 20 more in this area
Provider registry22
harness.provider_registry.provider_registry_get_providerGet providerreadharness.provider_registry.provider_registry_list_provider_versionsList provider versionsread- and 20 more in this area
Artifacts21
harness.artifacts.artifacts_list_artifact_digestsList an artifact target's digestsreadharness.artifacts.artifacts_list_artifact_variantsList an artifact's scanned versionsread- and 19 more in this area
Executions21
harness.executions.executions_get_resource_change_detail_by_addressGet resource change detail by addressreadharness.executions.executions_list_cost_changes_all_stagesList cost changes for all stagesread- and 19 more in this area
Labels21
harness.labels.list_repo_label_valuesList label values at repo levelreadharness.labels.list_repo_labelsList labels at repo levelread- and 19 more in this area
Remediation21
harness.remediation.check_artifact_and_deploymentsCheck Artifacts And Deploymentsreadharness.remediation.get_artifact_list_for_remediationGet Artifact List for Remediationsread- and 19 more in this area
Variable sets21
harness.variable_sets.variable_sets_list_variable_set_references_account_levelGet Variable Set References list on account levelreadharness.variable_sets.variable_sets_list_variable_set_references_org_levelGet Variable Set References list on org levelread- and 19 more in this area
Budgets20
harness.budgets.get_accessible_budget_summariesFilter budgets with RBAC checksreadharness.budgets.get_budget_summary_metricsBudget count metrics with RBAC checksread- and 18 more in this area
Filter20
harness.filter.ccmget_filter_listList Filtersreadharness.filter.get_filter_listList Filtersread- and 18 more in this area
Templates20
harness.templates.get_batch_templatesBatch fetch multiple templatesreadharness.templates.get_refreshed_yamlGet YAML with updated Template Inputsread- and 18 more in this area
User20
harness.user.check_if_last_adminCheck if user is last adminreadharness.user.get_aggregated_usersGet list of usersread- and 18 more in this area
User group20
harness.user_group.get_batch_users_group_listList User Groups by filterreadharness.user_group.get_filtered_user_groups_listGet filtered User Groupsread- and 18 more in this area
Authentication settings19
harness.authentication_settings.get_authentication_settingsGets authentication settings for the given Account IDreadharness.authentication_settings.get_authentication_settings_v2Gets authentication settings version 2 for the given Account IDread- and 17 more in this area
Ldap settings19
harness.ldap_settings.get_account_ldap_settingsGet ldap settings for the accountreadharness.ldap_settings.validate_connection_settingsValidate Ldap Connection Settingsread- and 17 more in this area
Notification rules19
harness.notification_rules.list_notification_rulesList Notification rulesreadharness.notification_rules.list_notification_rules_accountList Notification rules at account levelread- and 17 more in this area
Orchestration executions19
harness.orchestration_executions.get_orchestration_execution_activity_inputsGet Activity Execution Inputsreadharness.orchestration_executions.get_orchestration_execution_activity_outputsGet Activity Execution Outputsread- and 17 more in this area
Notification channels18
harness.notification_channels.get_notification_channelGet Notification channelreadharness.notification_channels.get_notification_channel_accountGet Notification channelread- and 16 more in this area
Pipeline execution details18
harness.pipeline_execution_details.get_execution_urlFetch Execution Urlreadharness.pipeline_execution_details.get_list_of_execution_identifierList Execution Identifierread- and 16 more in this area
Services 18Test targets 18Agents 17Clusters 17Entities 17Registry V3 versions 17Connectors 16Webhook 16Environments 15Onboarding 15Pipeline 15Recommendations 15Rule 15Pipeline input set 14Registry V2 14Registry V3 firewall 14Spaces 14Triggers 14Auto stopping rules 13Dashboards 13File store 13Rule executions 13Pipeline step 12Pipelines 12Recommendation jira 12Registry V3 lifecycle 12Releases 12Secrets 12Settings 12Usage 12Variables 12Backstage env variable 11Commitment orchestrator setup ap is 11Fault templates 11Orchestration processes 11Overrides 11Probes 11Project mappings 11Risks 11Account template 10Application set 10Experiment templates 10Freeze crud 10Org template 10Organization 10Project template 10Registry V3 packages 10Role assignments 10Rules 10Scans 10Downtime 9External data provider 9Infrastructure 9Input sets 9Integration 9Orchestration activities 9Pipeline execute 9Registry V3 registries 9Rule enforcement 9SL OS dashboard 9System 9Action templates 8Actions 8Budget groups 8Catalog custom properties 8Chaos recommendations 8Cost categories 8Data job status 8Delegate setup resource 8Faults 8Feature flags 8Governance overview 8Infrastructures 8Integrations 8Ip allowlist 8Perspectives 8Plugin info 8Policies 8Registry V3 signing keys 8Release groups 8Replication 8Service account 8Streaming destinations 8Token 8Webhooks 8Account banner 7Account services 7Accounts 7Aggregation rules 7366 smaller areas 1,196
This index shows a sample of each area. All 2,994 actions, with their inputs and prices, are in the catalogue, which a free sandbox key can search.
Questions
The Harness Platform MCP server, answered.
Is there a Harness Platform MCP server for Claude?
Yes. actionlake is an MCP gateway: a remote MCP server that gives Claude, or another MCP client, 2,994 Harness Platform actions across ansible, pullreq, repository and workspaces, through the same eight tools it uses for every other connector. actionlake is independent; it is not made or endorsed by Harness Platform.
What do I need to connect Harness Platform?
Harness API key token: in Harness open your profile (or Account Settings > Access Control > Service Accounts),…. Plus your account id and your host. The free call account { op: "connections.requirements", service: "harness" } lists exactly what to paste.
Can an AI agent delete or change things in Harness Platform?
Only if its key allows it, and only with an explicit confirmation. 357 Harness Platform actions are destructive, including delete a Role, soft delete repository and delete a project. Each runs only with confirm: true. Under the human confirm policy, or an approval rule on the key, the run waits until a named person approves it.
Will an AI agent use up my Harness Platform API limit?
actionlake paces every call to the limits in the Harness Platform service file: 300 calls a minute per connection. Repeated reads are answered from cache (1,288 Harness Platform actions can be), which never reaches Harness Platform.
What does a Harness Platform action cost?
One upstream call, the only unit actionlake bills. A list that pages bills one call per page (2,994 Harness Platform actions page). Cache hits, searches, dry runs and calls Harness Platform refused with 401, 403, 429 or a 5xx are free. The Free plan includes 100,000 calls a month with no card. Paid plans start at US$49 a month for 250,000 calls, with overage from US$0.30 down to US$0.18 per 1,000.
Which Harness Platform API does it cover?
The Harness Platform: CI/CD pipelines, GitOps, Code, Artifact Registry, Feature Flags, Chaos, Cloud Cost, IaCM, STO, SRM and platform administration API Harness APIs (OpenAPI 3.0.3, spec 1.0) at app.harness.io by default, or the base URL set when you connect. It is built from Harness Platform's documentation as read on 23 October 2026. 2,994 actions in 485 areas, each one in the signed service file (version 1.0.0).
Does actionlake keep my Harness Platform data, and where?
actionlake processes and stores data in Sydney, Australia; there is no US region yet. Reads may be cached for up to 5 minutes, and a write clears the reads it affects. Each run's request and response is kept, sealed, for your plan's trace window: 1 day on Free, 3 days on Starter, 7 days on Pro, 14 days on Team, 30 days on Business and 30 days on Enterprise. Your Harness Platform credential is sealed under your tenant's key, never logged and never returned. Your traffic is not used to train models.
Which AI clients can use it?
Claude Code, tested by us end to end, and the Claude app as a custom connector, tested by us with an existing key. Other MCP clients that support remote servers over Streamable HTTP (MCP revisions 2025-06-18 and 2025-11-25) and OAuth dynamic client registration, or a bearer header, should connect the same way; we have not tested them. Or plain REST, under the same action names.
Give your agent Harness Platform today.
Start free - 100,000 calls/month, no card. No signup. Hard caps on every key. Cached answers are free.