scylladb

Generate standardized commit messages following the Conventional Commits specification. Use when asked to commit changes, write a commit message, create a conventional commit, or when committing code. Analyzes staged changes and produces properly formatted commit messages.

Create a new implementation plan or sub-plan for cqlsh-rs features, refactoring, or infrastructure work. Use when asked to plan a feature, create a design document, write an implementation plan, break down a task into phases, or design architecture for a component. Produces structured, AI-executable plans with deterministic language.

Guide the end-to-end development process for cqlsh-rs features: review plans, design tests, implement code, write tests, and update plan documents. Use when starting a new feature, picking up the next development task, or following the project's development workflow from plan to implementation.

Implement and configure AI-powered CI failure analysis workflows for GitHub Actions. Use when setting up CI failure summaries, configuring claude-code-action for failure diagnosis, writing workflow_run triggered analysis, creating collapsed PR comments with failure diagnostics, implementing flaky test detection, or working on recurring issue tracking. Covers the full SP15 plan.

Author and maintain GitHub Actions workflows for CI/CD pipelines. Use when creating new workflows, modifying ci.yml, adding workflow jobs, configuring matrix builds, setting up caching, managing secrets, writing workflow_run triggers, creating release pipelines, or debugging GitHub Actions issues. Covers CI, benchmarking, release, and documentation workflows for cqlsh-rs.

Run Clippy with strict lint settings and fix warnings for Rust code. Use when asked to lint code, fix clippy warnings, enforce Rust idioms, check code quality, or run static analysis. Applies project-specific lint configuration and explains fixes.

Generate comprehensive Rust tests for cqlsh-rs modules. Use when asked to write tests, add test coverage, generate unit tests, create integration tests, or improve test coverage for Rust code. Orchestrates test creation following project conventions, cargo test patterns, and the cqlsh-rs testing strategy.

Save lessons learned, debugging insights, and project-specific knowledge into persistent memory files. Use when asked to remember something, save a lesson, record a gotcha, document a pattern, or persist knowledge for future sessions. Syntax: /remember [domain] lesson content.

Guides AI-assisted code review of SCT pull requests. Use when reviewing a PR, checking a diff for correctness, evaluating method signature changes across class hierarchies, verifying override compatibility, checking import conventions, error handling patterns, backend impact, test coverage, or provision label requirements. Covers inheritance safety, polymorphic method audits, and SCT-specific review criteria.

Guides writing remote package installation commands (apt-get, yum, dnf, zypper) in SCT code. Use when adding apt-get install/update, yum install, dnf install, or zypper install calls via remoter.run/remoter.sudo. Ensures timeouts, retries, and non-interactive flags are always present to prevent hangs in CI.

Generate weekly commit summary reports for SCT repository. Use when asked to create a commit summary, weekly report, changelog, or "last week in SCT" issue. Applies to summarizing git commits from scylla-cluster-tests master branch for developer audiences. Covers running sct_commits_summary.py, filtering commits by importance, and writing prose summaries with embedded GitHub links.

Guides writing and debugging unit tests for the SCT framework using pytest conventions. Use when creating new test files in unit_tests/, adding test cases, mocking external services, setting up fixtures, or reviewing test coverage. Covers network-blocking patterns, FakeRemoter, moto for AWS mocking, monkeypatch, and common pitfalls.

Guides writing and debugging integration tests for the SCT framework that interact with real external services. Use when creating tests requiring Docker, AWS, GCE, Azure, OCI, or Kubernetes backends. Covers service labeling, credential skip patterns, Docker Scylla fixtures, resource cleanup, and common pitfalls.

Fix inline merge conflict markers in backport PRs by resolving conflicts and recommitting cleanly with original metadata preserved. Use when a backport PR has unresolved conflict markers, a cherry-pick produced merge conflicts, or a PR has the 'conflicts' label and needs to be made ready for review. Supports bulk mode for multiple PRs.

Guides writing new nemesis (chaos engineering disruptions) for the SCT framework. Use when creating a new NemesisBaseClass subclass, adding disruption logic, setting nemesis flags, or configuring target node pools. Covers the sdcm/nemesis/ package structure, auto-discovery, flag filtering, CI configuration, and unit testing patterns.

Use when asked to generate an implementation plan, draft a plan, save a plan, or design a feature rollout for the SCT repository. Supports two formats: full 7-section plans for multi-phase work (1K+ LOC, tracked in MASTER.md) and lightweight mini-plans for single-PR changes (under 1K LOC, stored in docs/plans/mini-plans/). Routes automatically based on PR plans label, user input, or task size estimate.

Multi-agent code review for Python projects. Use when reviewing local uncommitted changes, pull request diffs, or when asked to do a code review. Dispatches specialist agents (bug-hunter, security-auditor, test-coverage, code-quality, contracts, historical context) and aggregates results with confidence scoring.

Configures Python projects with modern tooling (uv, ruff, ty). Use when creating projects, writing standalone scripts, or migrating from pip/Poetry/mypy/black.

Guides the design and structuring of workflow-based Claude Code skills with multi-step phases, decision trees, subagent delegation, and progressive disclosure. Use when creating skills that involve sequential pipelines, routing patterns, safety gates, task tracking, phased execution, or any multi-step workflow. Also applies when reviewing or refactoring existing workflow skills for quality.

Guides refactoring of Python test suites to reduce duplication using pytest.mark.parametrize, split large monolithic test files into focused modules, and deduplicate mirrored sync/async test classes. Use when test files exceed 400 lines, when multiple test functions share identical structure with different inputs, or when sync and async test classes are copy-pasted mirrors of each other.

Guides the creation and structuring of project plans in docs/plans/. Use when writing feature-parity plans, gap analyses, documentation plans, or implementation roadmaps. Covers plan templates, status tracking conventions, phase numbering, task tables, and test plan sections.

Guide for running and writing coodie vs cqlengine performance benchmarks. Use when asked to run benchmarks, add new benchmarks, investigate performance, or debug benchmark failures.

Guides ScyllaDB Cloud users through implementing and optimizing Vector Search for semantic similarity, RAG, and similar use cases. Use this skill when users need to store and query embeddings, build vector indexes (HNSW), run approximate nearest neighbour (ANN) queries, apply filtering (global/local secondary indexes), configure quantization, or integrate with LLM frameworks (e.g. LangChain). Also use when users mention "vector", "embeddings", "similarity search", "ANN", "nearest neighbor", "RAG", "semantic search", or "recommendation system" in the context of ScyllaDB.

Guide users through connecting to a ScyllaDB Cloud cluster. Use this skill when a user needs to connect to ScyllaDB Cloud, configure driver credentials, or troubleshoot connection issues. Triggers on "connect to ScyllaDB Cloud", "ScyllaDB connection", "ScyllaDB driver setup", "CQL connection", "DC-aware load balancing", "ScyllaDB credentials", "connection bundle".

ScyllaDB CQL data modeling patterns and anti-patterns. Use when designing tables, reviewing schemas, migrating from SQL or MongoDB, or troubleshooting performance issues caused by schema problems. Triggers on "design schema", "ScyllaDB data model", "partition key", "clustering column", "primary key", "CQL table design", "ALLOW FILTERING", "large partitions", "hot partitions", "query-first design", "secondary index", "materialized view", "schema review", "wide partition", "tombstones", "time series".

Investigate failed e2e tests from Ginkgo JSON reports and must-gather artifacts, systematically analyzing logs, events, and resource states to identify root causes.

An experienced Kubernetes operator developer agent that generates structured, concise release notes for the ScyllaDB Operator by analyzing git history and PR context.

Guides the design and structuring of AI agent skills for the Argus repository with multi-step phases, progressive disclosure, and dual-platform compatibility (GitHub Copilot and Claude Code / OpenCode). Use when creating new skills, reviewing existing skills, restructuring AI guidance into modular skill directories, editing SKILL.md files, or improving agent instructions.

Use when asked to generate an implementation plan, draft a plan, save a plan, or design a feature rollout for the Argus repository. Supports two formats: full 7-section plans for multi-phase work (1K+ LOC, tracked in MASTER.md) and lightweight mini-plans for single-PR changes (under 1K LOC, stored in docs/plans/mini-plans/). Routes automatically based on PR plans label, user input, or task size estimate.

Review changes to the new or already existing integration tests from `*_integration_test.go` files. Do this before making the changes to integration tests yourself or when asked to review them.