LLM 與 Agent

Build timeout-resistant Claude Code workflows with chunking strategies, checkpoint patterns, progress tracking, and resume mechanisms to handle 2-minute tool timeouts and ensure reliable completion of long-running operations.

Generate LLM-backed release notes between git tags using the repo release-scripts. Use when asked for changelogs, release notes, or summaries between two tags.

Complete 6-step self-healing development loop per .council/AGENTS.md. Use for any feature implementation from start to finish.

Эксперт по оркестрации AI агентов. Используй для multi-agent systems, agent coordination, task delegation и agent workflows.

Manage the PR review feedback loop: monitor CI checks, fetch review comments, and iterate on fixes.Use when: (1) pushing changes to a PR and waiting for CI/reviews, (2) user says "new reviews available",(3) iterating on PR feedback from Gemini, Cursor, Claude, or other reviewers, (4) monitoring PR status.Supports multiple review bots: Gemini Code Assist, Cursor Bugbot, and Claude agent fallback.Automatically detects priority levels from different bot formats and handles rate limits.RECOMMENDED: Spawn a Task agent (subagent_type: general-purpose) to execute the review loop autonomously.See "Recommended Usage: Run as Task Agent" section for the prompt template.CRITICAL: When using this skill, NEVER use raw git commit/push commands. ALWAYS use commit-and-push.sh script.The user has NOT granted permission for raw git commands - only the script is allowed.

Validate that curriculum fully covers required standards, check depth and rigor of coverage, verify assessment alignment, and generate coverage reports for administrators. Use for comprehensive validation. Activates on "validate coverage", "standards compliance", or "coverage report".

Fetch and display filtered X (Twitter) timeline based on your interests. Shows AI/ML research, developer tools, and technical content while filtering out noise, rage bait, and irrelevant posts. Uses Claude Haiku to score relevance against your objectives.

Defines the autonomous development workflow. Use when orchestrating work between agents, managing issue lifecycles, or understanding the development loop.

Markdown documentation standards for LLM-optimized architecture including section numbering rules for Pandoc PDF generation. Use when writing markdown documentation, creating skills, or authoring content that may be converted to PDF.

MANDATORY protocol for parallel agent execution when facing 3+ independent failures or tasks that can be investigated without shared state or dependencies - dispatches multiple Claude agents to investigate and fix independent problems concurrently

Codex handoff checklist for agents.

Configure, manage, update and review Claude Code permissions, sandboxing, and tool access. Use when user wants to set up permissions, configure sandboxing, update allowed tools, manage settings.json permissions, or review permissions in skills or commands or agents or settings.json. When user writes a new skill, command, agent, or updates settings.json, they should use this skill to manage permissions.

Use when planning software development tasks, before writing code or implementation plans - refines rough technical ideas into fully-formed designs through collaborative questioning, alternative exploration, and incremental validation. For software architecture, components, data flow, and technical design decisions. Don't use during clear 'mechanical' coding processes

Persist task state across sessions using .claude/ files. Resume work with 'continue'. Never use TodoWrite - use tasks.md, requirements.md, session.md instead.

Workflow for processing large Things3 inboxes (100+ items) using LLM-driven confidence matching and intelligent automation. Integrates with personal taxonomy and MCP tools for efficient cleanup with self-improving pattern learning.

Use when implementing new features. Enforces TDD workflow - write tests FIRST, then implementation. Ensures AAA pattern, proper coverage, and quality test design.

Complete product launch workflow coordinating 15+ specialist agents across research, development, marketing, sales, and operations. Uses sequential and parallel orchestration for 10-week launch timeline.

Automatically retrieve relevant information from the organization's CustomGPT.ai knowledge base when answering questions about documented topics, policies, procedures, or technical specifications.

This skill should be used when the user wants to adjust Claude's interaction style and personality level. It provides four distinct tone modes ranging from formal professional to casual buddy, allowing users to customize the interaction experience based on their mood, task type, or preference. Trigger when user requests a specific vibe, tone, or personality change.

Expert assistant for Bible reference lookups and RPC function usage in the KR92 Bible Voice project. Use when fetching Bible verses, generating verse lookups, validating references, working with OSIS format, or choosing optimal lookup patterns.