datum-platform

Display user correction statistics and patterns. Shows recent corrections, their types, affected agents, and emerging patterns from user feedback.

Trigger deployment workflow for a feature. Invokes the SRE agent to handle deployment configuration, staging rollout, and production promotion.

Quick-start feature discovery. Creates a pipeline entry and invokes the product-discovery agent to begin the discovery process.

Analyze review findings and session learnings to extract patterns, update the pattern registry, and automatically promote high-confidence patterns to agent runbooks. Run periodically to keep runbooks current.

Display pattern statistics from the learning engine. Shows top patterns, their frequency, trends, and which agents are affected.

Orchestrate the Datum Cloud feature development pipeline. Use to start new features, check status, advance to the next stage, or route work to the appropriate agent.

Invoke code review on the current branch or a specific feature. Runs the code-reviewer agent with appropriate context from the pipeline.

Show pattern trends over time and generate alerts for concerning changes. Helps identify areas where the team is improving or struggling.

MUST BE USED for Go backend implementation on Kubernetes aggregated API servers. Use for implementing API types, storage backends, server configuration, request handlers, code generation, and any Go code in cmd/, internal/, or pkg/ directories. Use for running code generation, linting, and testing. Use AFTER architect has produced a design.

MUST BE USED as a post-implementation quality gate. Use after api-dev, frontend-dev, or sre complete changes. Use when someone says "review this" or "check this code" or "is this ready to merge." Reviews for correctness, security, convention compliance, and platform integration completeness. Read-only — produces findings, never modifies code.

MUST BE USED for React/TypeScript UI implementation, component development, page layouts, form implementation, dashboard creation, and any frontend code changes. Use for building admin UIs, service provider portals, consumer-facing resource management, or any browser-based tooling.

Use for architectural design and planning of Kubernetes platform features. Produces design documents for aggregated API servers, controllers, event processing pipelines, and platform capabilities. Use BEFORE api-dev or other implementation agents. Outputs designs to .claude/pipeline/designs/.

MUST BE USED for Kustomize manifests, CI/CD pipeline configuration, Dockerfile changes, RBAC configuration, observability setup, deployment configuration, container security, TLS configuration, FluxCD deployment, OCIRepository configuration, and any infrastructure-as-code changes. Use for anything in config/, .github/, Dockerfile, or infra repository changes.

Use for API documentation, onboarding guides, runbooks, configuration references, migration guides, changelogs, release notes, README updates, inline code documentation, and any user-facing or operator-facing written content that lives in docs/ or README files. Use when someone says "document this" or "write a guide for" or "update the README."

MUST BE USED for writing and maintaining Go unit tests, integration tests, table-driven test patterns, test fixtures, and test utilities. Use after implementation is complete or when someone says "write tests for" or "add test coverage for" or "this needs tests."

Covers creating C4 architecture diagrams using PlantUML. Use when documenting system architecture, component relationships, or container layouts for technical documentation.

Covers activity timeline integration using the Activity system. Use when implementing ActivityPolicy resources, emitting Kubernetes events for activity tracking, or exposing activity timelines to users via CLI, UI, or API.

Decision framework for determining which platform capabilities a feature needs (Quota, Insights, Telemetry, Activity). Use when planning a new feature to assess which capabilities apply.

Covers proactive issue detection using the Insights system. Use when implementing InsightPolicy resources with CEL-based rules to automatically detect misconfigurations, health issues, or security concerns.

Covers resource quota integration using the Milo quota system. Use when implementing ResourceRegistration, ClaimCreationPolicy, or GrantCreationPolicy resources for quota enforcement and allocation.

Covers observability integration including metrics, traces, and logs following OpenTelemetry conventions. Use when instrumenting services for monitoring and debugging.

Covers commit message conventions including format, structure, and content guidelines. Use when writing commits to ensure consistent, meaningful commit history across Datum Cloud repositories.

Covers patterns for building Kubernetes controllers using controller-runtime with Milo's multi-cluster runtime provider. Use when implementing controllers that need to operate across project control planes.

Covers CI/CD patterns using GitHub Actions with reusable workflows. Use when setting up pipelines, Taskfiles, or Dockerfiles for Datum Cloud services.

Trigger deployment workflow for a feature. Invokes the SRE agent to handle deployment configuration, staging rollout, and production promotion.

Covers the design token architecture for Datum Cloud UIs including colors, spacing, typography, and theme support. Use when implementing UI components to ensure visual consistency.

Quick-start feature discovery. Creates a pipeline entry and invokes the product-discovery agent to begin the discovery process.

Analyze review findings and session learnings to extract patterns, update the pattern registry, and automatically promote high-confidence patterns to agent runbooks. Run periodically to keep runbooks current.

Covers GitOps deployment using FluxCD with OCIRepository and Flux Kustomization resources. Use when configuring deployments, dependency ordering, or inline patching in the infra repository.

Covers Go coding conventions including import grouping, error handling, naming, and comments. Use when writing or reviewing Go code to ensure consistency across services.

Covers patterns for building Kubernetes aggregated API servers including REST handlers, storage interfaces, scheme registration, and type definitions. Use when implementing API types, storage backends, or server configuration.

Covers Kustomize deployment patterns using base, components, and overlays structure. Use when organizing Kubernetes manifests for different environments.

Provides automatic pattern extraction and knowledge accumulation from review findings. Use when running /evolve, /patterns, or /trends commands, or when understanding how patterns are promoted to runbooks.

Covers the Identity and Access Management system including ProtectedResource, Role, and PolicyBinding definitions. Use when implementing authorization for new resource types or defining service roles.

Display pattern statistics from the learning engine. Shows top patterns, their frequency, trends, and which agents are affected.

Covers pipeline orchestration for feature development including stages, handoffs, and artifact management. Use when coordinating work through the discovery to deployment pipeline.

Orchestrate the Datum Cloud feature development pipeline. Use to start new features, check status, advance to the next stage, or route work to the appropriate agent.

Provides high-level understanding of the Datum Cloud platform architecture. Use when onboarding to the codebase, understanding the system structure, or learning about multi-tenancy, resource hierarchy, and service architecture.

Covers pull request description conventions including structure, sections, and content guidelines. Use when creating PRs to ensure consistent, informative descriptions across Datum Cloud repositories.

Invoke code review on the current branch or a specific feature. Runs the code-reviewer agent with appropriate context from the pipeline.

Explains the runbook system for accumulating learnings including patterns, anti-patterns, and service-specific guidance. Use when understanding how agents use runbooks or when manually adding entries.

Show pattern trends over time and generate alerts for concerning changes. Helps identify areas where the team is improving or struggling.

Captures and learns from direct user corrections to Claude Code's outputs. Complements the code-reviewer-driven learning system with user-driven feedback for continuous improvement.

1 hook across 1 event