opendatahub-skills

[DEPRECATED] Backwards-compatibility umbrella for the ODH AI Helpers plugins. Re-exports the skills that existed before the split into the individual odh-* plugins under their original odh-ai-helpers:* names, so existing agents and workflows keep working. Install the odh-* plugins you need, then uninstall this one; it will be removed after the migration window.

72 skills

Documentation review, writing, and workflow tools for AsciiDoc and Markdown documentation. Includes an orchestrated multi-step pipeline, standalone review skills, codebase analysis for onboarding, and JIRA/PR integration.

52 skills · 16 agents

Everything you need for PatternFly development -- React components, design guidance, migration, and MCP docs. Installs all PatternFly sub-plugins in one step.

55 skills

Consensus-based security review for RHOAI strategy documents (STRATs). An orchestrator spawns three independent reviewers to identify security risks, then synthesizes findings with confidence tagging based on cross-reviewer agreement. Covers 39 catalog patterns across auth, data protection, cryptographic compliance, network security, supply chain, and infrastructure.

29 skills

React component development -- coding standards, testing, and structure

16 skills · 3 agents

Claude Code plugin with specialized skills for DevOps and cloud-native development workflows. Provides skills for GitLab CI/CD analysis, AWS CloudWatch Logs troubleshooting, Slack integration, GitLab branch management, Jira utilities, Konflux ITS analysis, cloud infrastructure provisioning via mapt, and Google Workspace integration.

18 skills

End-to-end test planning workflow for RHOAI: generate E2E/UI-focused test plans from Jira strategies, create traceable test cases, implement executable automation code, verify UI tests against live clusters via Playwright, publish to GitHub, resolve review feedback, and score plans with deterministic evidence gates and automated rubrics.

16 skills

Claude Code skills for creating, reviewing, splitting and submitting work items — RFEs to the RHAIRFE Jira project and Initiatives to RHAISTRAT — through one generic pipeline (/rfe-create, /rfe-review, /rfe-split, /rfe-submit, /rfe-auto-fix, /rfe-speedrun; --type initiative for Initiatives). The legacy dotted names (/rfe.create and friends) remain as compatibility aliases.

13 skills · hooks

Jira ticket management, search, triage, and automation

13 skills

Claude Code skills for creating, reviewing, and submitting strategies to the RHAISTRAT Jira project. Provides an automated pipeline from initial creation through refinement, adversarial review with independent reviewers, and human sign-off workflow with pull/push/signoff gates.

12 skills · hooks

Python package analysis, security auditing, and build complexity assessment

11 skills · 1 agent

AI skills for Python package onboarding into the RHAI distribution pipeline. End-to-end automation covering packaging investigation, license checking, security auditing, build failure analysis, fondue monorepo onboarding, constraint bug descriptions, hardware variant resolution, probe test creation, Jira context summarization, and executive summary generation. Designed to run inside a Claude Code container as part of the package-onboarding CI pipeline.

11 skills · hooks

Generic agentic evaluation for skills and agents. Provides end-to-end skills to analyze, test, score, review, and iteratively improve agent skills, plus compare models/configurations and run Design-of-Experiments (ANOVA) sweeps. MLflow support for experiment tracking, tracing, and reporting. Schema-driven evaluation via eval.yaml with support for inline, LLM-based, and external judges.

10 skills · hooks

AsciiDoc documentation generation, validation, review, and ADR review

9 skills

vLLM backport triage, cherry-pick automation, and requirements comparison

8 skills

Design audit -- validate existing code and designs against PatternFly standards

7 skills · 1 agent

Claude Code plugin for the Jira autofix pipeline. Provides orchestrator skills, agent prompt files, and deterministic Python scripts for automated bug fixing, CVE remediation, and ticket triage. Designed to run inside a Claude Code container as part of a CI pipeline.

6 skills · hooks

Quality tooling and automation for RHOAI component development. Includes automated repository analysis, build validation, and test pattern extraction.

7 skills

DevOps and TestOps automation skills for ODH/RHOAI — component onboarding, Konflux CI/CD, release management, delivery pipelines, and operational tooling.

6 skills

Assess RFEs against quality criteria using a structured rubric.

3 skills · 2 agents · hooks

Git utilities, GitHub/GitLab workflow automation, and CI debugging

6 skills

Execute RHOAI SPIKE investigations with human-in-the-loop approval gates. 9-step lifecycle: intake, plan, Jira sync, AI research enrichment with hallucination detection, pytest test suites on OpenShift, rubric-based scoring with security gates, and RFE input generation. Supports both runtime and protocol library assessment.

2 skills · 4 MCP servers

Autonomous knowledge management skills for keeping AI context files (CLAUDE.md, AGENTS.md) up to date. Scans merged PRs, extracts relevant knowledge using parallel agents, and proposes updates as a git-apply-able patch for human review. Supports GitHub and GitLab.

5 skills

Code review and quality -- adversarial review, security patterns

5 skills

Design guide -- component selection, interaction patterns, AI experience patterns

3 skills · 1 agent

Assess RHAISTRAT strategies against quality criteria using a scored rubric with calibration examples. Scores across four dimensions: feasibility, testability, scope, and architecture.

2 skills · 1 agent

AI skills for AutoQA CI/CD test failure analysis and triage. Covers root cause analysis of test failure logs, matching failures against historical Jira tickets, and classifying failures as known infrastructure false alarms. Designed to run inside a Claude Code container as part of the AutoQA CI pipeline.

3 skills

AI code review skills. A single-pass GitLab merge request reviewer that posts inline comments and a summary (in CI) or displays them locally, a risk-tiered review panel with one reviewer subagent per lens and skeptic subagents that verify each finding, and a security-only diff review with a false-positive filter per finding.

3 skills

Gmail, Google Calendar, Docs, and Drive integration

3 skills

Konflux application and component management

3 skills

ODH module operator scaffolding, migration, and compliance checks

3 skills

PyTorch cross-language analysis with TorchTalk

3 skills

Team weekly reports, engineer activity snapshots, and delivery postmortems

3 skills

Accessibility auditing, reporting, and documentation

3 skills

PF version migration -- breaking change detection, class scanning, upgrade planning

3 skills

Pipeline failure analysis skills for AIPCC CI/CD pipelines. Collects structured wheel failures into one audit report, groups failed jobs by shared root cause, and performs root cause analysis with structured findings and confidence-rated diagnoses. Includes deterministic CI helpers and analysis skills used by pipeline-failure-analyzer.

3 skills

Decompose RHAISTRAT strategies into implementation epic dependency graphs with adversarial review, bounded revision and AI implementability scoring. Accepts strategy IDs or a JQL query for non-interactive batch processing.

1 skill · hooks

CodeRabbit review triage and project-conformant unit test generation

2 skills

llm-d release orchestration for opendatahub-io

2 skills

RPM build failure analysis and non-Red Hat RPM detection

2 skills

Supply-chain security alerting and OCI image CVE comparison

2 skills

Score a repository's readiness for disconnected / air-gapped OpenShift deployments. Scans for image manifest completeness, digest enforcement, runtime egress, and Python dependency validation. Supports automatic detection of image management patterns (env var vs static CSV) and cross-references against the opendatahub-operator manifest.

1 skill

Inspect Enterprise Contract CVE scan results from Konflux-built container images and test ECP exception removal. Extracts the full Clair REPORTS data from cosign attestations (the same data EC's cve.cve_blockers rule evaluates), supports human-readable and JSON output, and can drive local or cluster-based EC policy validation to check whether a cve.cve_blockers exception is still needed.

1 skill

General-purpose helpers and learning mode

1 skill

MaaS nightly QE impact analysis

1 skill

PatternFly MCP server -- component documentation, design token lookup, and accessibility guidance via the Model Context Protocol

1 MCP server