LLM & Agents

Analyze existing PHPUnit test failure logs without running tests. Use when:- User says "fix the test failures" (after manually running tests)- User says "what tests are failing?"- User points to specific log file- Tests were run outside Claude's contextDelegates to fixer agent (sonnet) to find logs, parse failures, and implement fixes.Does NOT execute tests - use phpunit-runner for that.

GitHub issue情報を取得し、対応用のタスクファイル作成をサポートします。スクリプトがissue情報を取得し、Claude Codeがその情報を元にtmp/todoフォルダにタスクファイルを作成します。

This skill should be used when the user is stuck in a git-hex-driven rebase, merge, or cherry-pick due to conflicts and wants the agent to inspect, resolve, and then continue or abort safely. Trigger phrases include: "resolve conflicts", "rebase is stuck", "conflict markers", "continue the rebase", "abort the rebase".

Parameter-efficient fine-tuning with LoRA and Unsloth. Covers LoraConfig,target module selection, QLoRA for 4-bit training, adapter merging, andUnsloth optimizations for 2x faster training.

Use when creating or developing, before writing code or implementation plans - refines rough ideas into fully-formed designs through collaborative questioning, alternative exploration, and incremental validation. Don't use during clear 'mechanical' processes

Guide for building applications on Cloudflare's edge platform. Use when implementing serverless functions (Workers), edge databases (D1), storage (R2, KV), real-time apps (Durable Objects), AI features (Workers AI, AI Gateway), static sites (Pages), or any edge computing solutions.

Search for negative news coverage, controversies, and reputational risks associated with individuals or companies across news sources and media databases

Techniques for ensuring LLM responses adhere to strict JSON schemas, utilizing Pydantic models, JSON mode, and schema-based refusals. Triggers: structured-output, pydantic, json-schema, json-mode, llm-response-parsing.

Ensures proper cloud storage operations with retry logic, error handling, streaming, and efficient I/O patterns. Activates when users work with object_store for S3, Azure, or GCS operations.

Helix Personal AI Infrastructure for Sean Sylvia. CORE is the foundational skillthat loads at every session. Orchestrates specialized agents for health economicsresearch, grant writing, VBC modeling, and knowledge management. Spawns subagentsfor complex tasks and manages persistent memory, projects, and Obsidian vaultsynchronization.

Query remote Parquet files via HTTP without downloading using DuckDB httpfs. Leverage column pruning, row filtering, and range requests for efficient bandwidth usage. Use for crypto/trading data distribution and analytics.

Use when animating multi-part objects, character appendages, fabric, hair, or any motion requiring realistic drag, momentum, and settling behavior.

Install Codex CLI MCP assistant. Use when user needs Codex CLI capabilities in their project.

Unified login wizard combining Claude API authentication with CIPS identity setup. Use when /login invoked, first-run detected, or identity reset requested. Follows @asking-users PARAMOUNT patterns.

End-to-end testing scenarios for Supabase - complete workflow tests from project creation to AI features, validation scripts, and comprehensive test suites. Use when testing Supabase integrations, validating AI workflows, running E2E tests, verifying production readiness, or when user mentions Supabase testing, E2E tests, integration testing, pgvector testing, auth testing, or test automation.

This skill should be used when the user asks to "create literate tests", "generate markdown tests", "specification-as-tests", "TDD with markdown", "agent-driven testing", or mentions test suites where markdown IS the test format. NOT for pytest/jest/unittest. Creates .md test files with inline assertions and uses a bundled custom test runner.

Create specialized infrastructure agent definitions for platform/service management (Grafana, Prometheus, Traefik, ERPNext, etc.). Use when the user requests creation of an agent for a specific technology platform or infrastructure component. This skill produces complete agent prompts with integrated research, SOPs, tool references, and handoff protocols following the Linear-First Agentic Workflow framework.

Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends Claude's capabilities with specialized knowledge, workflows, or tool integrations.

Multi-phase workflow orchestration pattern for main agents. Enables structured planning, delegation to sub-agents, parallel execution, and result aggregation.

Enforce ultra-conservative token usage by minimizing prompt exposure, delegating bulk work, and tracking quotas.