DOCSEARCHbyAlgolia
Submit Docs

Nullplatform Documentation

website·Indexed 31 days ago

https://docs.nullplatform.com

Documentation for Nullplatform, featuring guides on the Nullplatform Resource Name (NRN) hierarchical string system, AI-assisted action items for detection and resolution, and the np-governance agent plugin. Includes details on managing categories, ownership, triage workflows, and configuring notification channels for platform events.

Tokens
491.6K
Snippets
1.7K
Records
3.4K
Agent score
49%

What's inside Nullplatform

  1. Overview of agent-backed scopes

    Agent-backed scopes provide a flexible mechanism to implement custom scopes for deploying diverse technologies, including LLM agents, banking cores, mobile applications, static frontends, and PL/SQL. They can also be used to customize Kubernetes deployments at a micro level. Agent-backed scopes maintain full compatibility with traditional scope features, including parameters, approvals, and policies, and provide a seamless UI experience for end users.

  2. Overview of Nullplatform AI plugins

    Nullplatform AI plugins allow developers to operate the platform directly from coding assistants (such as Claude Code, Kiro, and Cursor) using plain language. The plugins handle querying entities, paginating results, and calling the appropriate API endpoints based on the user's description. These plugins are a prerequisite for using other AI Ops features.

  3. Overview of nullplatform Workflows

    nullplatform Workflows allow you to create automations and AI agents by describing them in plain language and refining them on a visual canvas. Workflows can be triggered on a schedule, by platform events, on demand, or via webhooks. Key features include:

    • Visual Editing: Access the editor via Platform settings → Workflow Editor to adjust nodes and run workflows.
    • Connectivity: Includes pre-built nodes for the data lake, action items, notifications, approvals, and nullplatform APIs, as well as support for custom code steps and HTTP calls.
    • AI Agent Integration: Steps can be assigned to AI agents for judgment-based decisions. Agents can use other workflows as tools.
    • Human-in-the-loop: Use checklists or action items to require human decision-making.
    • Version Control & Debugging: Every save creates a new version. Workflows must be explicitly ACTIVATED to run. Each run preserves inputs, outputs, and logs for every step to facilitate troubleshooting.

  4. Overview of nullplatform core features

    nullplatform provides a suite of interoperable tools for platform automation. Each feature can be used independently or combined into a larger operational process. The general development flow for all features is: install AI plugins in a coding assistant, describe requirements in plain language, review the result in the UI, and activate.

  5. Overview of the nullplatform Notifications system

    The nullplatform notifications system allows users to automate workflows, create audit trails, and receive real-time alerts for platform events, such as new deployments and parameter approvals. The system is fully configurable, allowing users to define the destination, content, and trigger conditions for each notification. It integrates with deployments, parameters, and telemetry to provide a centralized observation and control mechanism.

  6. Overview of nullplatform AI Ops

    AI Ops is a suite of nullplatform features designed to automate recurring operational work. Key capabilities include identifying outdated runtimes before they cause incidents, gating risky deployments until specific checks pass, and tracking migrations across multiple teams. All AI Ops processes run on the organization's own data and adhere to organization-defined rules.

  7. Overview of Nullplatform ready-to-use automation suites

    Nullplatform provides several open-source automation suites that run in its production environment. These suites are available in the nullplatform/workflows GitHub repository and typically include a 30-minute tutorial for setup.
    # Source for all ready-to-use suites
git clone https://github.com/nullplatform/workflows

  8. Overview of GitOps for infrastructure management

    GitOps is a methodology for managing infrastructure and applications using version-controlled, declarative configurations. This approach ensures that all changes are tracked, auditable, and easily reversible. Key tools used to automate and simplify cloud infrastructure management within a GitOps workflow include ArgoCD, Crossplane, and the nullplatform Terraform provider.

  9. Overview of the Services Infrastructure Catalog

    The Services catalog manages infrastructure components (such as databases, message queues, and caches) that applications consume. The system distinguishes between 'specifications' (templates defining available types) and 'services' (concrete provisioned instances). 'Links' are used to connect applications to these services, providing the necessary credentials and configuration.

  10. Overview of nullplatform Agent notification channels

    Agent notification channels allow you to trigger actions within your own infrastructure when specific platform events occur. When an event matches the channel's source and filters, nullplatform sends a notification to the nullplatform agent, which then executes a specified command on the host machine.

  11. Overview of nullplatform authentication methods

    nullplatform secures access using two primary authentication methods: standard username and password credentials, and SAML authentication for integration with external identity providers (IDPs).

  12. Overview of the nullplatform agent

    The nullplatform agent is a lightweight service that can be deployed in various environments, including local laptops or Kubernetes clusters. It acts as a secure bridge between your infrastructure and the nullplatform API, enabling automation of tasks, monitoring of scopes and services, triggering of actions, and management of event-based notifications without requiring direct external exposure of your infrastructure. The agent is designed with a privacy-first approach where data remains within your cluster, and only minimal metadata and telemetry are transmitted to nullplatform.