AI Engineering

Clinical Decision Support Documents generates publication-ready clinical decision support reports for pharmaceutical and research settings. Use it for cohort analyses, evidence-graded treatment recommendations, and biomarker-driven decision algorithms.

Coding Tutor creates personalized coding tutorials that build on your existing knowledge and use your actual codebase for examples. It maintains a persistent learning trail with spaced repetition and quizzes.

Comprehensive Research Agent provides structured protocols for source validation, error recovery, and transparent reasoning in multi-step research tasks. Use it when gathering information from multiple web sources or running multi-tool research workflows to avoid silent failures and premature conclusions.

Vibe governs complex agent tasks by freezing requirements, bounding execution, and enforcing verification and phase cleanup. Use it when you explicitly invoke $vibe or need disciplined multi-skill orchestration with hard stops and proof artifacts.

J-Space establishes and operates the model's inner workspace for tasks that require multi-step reasoning, planning, long-horizon work, complex debugging, or calibrated confidence. Use it as an entry point that classifies the task and routes to the appropriate module.

Vibe Agents generates AGENTS.md and tool-specific AI configuration files for coding assistants using PRD and tech design documents. Use it to set up agent instructions before vibe-coding an MVP.

Vibe Research guides users through a structured deep research and market validation process for app ideas. Use it when starting a new project, validating an idea, or when the user asks to research or validate an app concept.

Vibe Tech Design guides users through creating a Technical Design Document for their MVP. Use it to plan architecture, choose a tech stack, and document key technical decisions.

CCS Delegation routes deterministic coding tasks to cost-optimized models via CCS CLI. Use it to auto-select glm, kimi, or custom profiles, enhance prompts with context, and execute or continue sessions for typos, tests, and refactors.

Skill Creator provides guidance for creating effective AgentSkills, including structure, scripts, references, and assets. Use it when designing, structuring, or packaging skills with scripts, references, and assets.

Agent Teams coordinates multiple Claude Code sessions as a team with shared task lists, mailbox messaging, and file-lock claiming. Use it to parallelize work on non-overlapping files or delegate cross-layer changes across a lead and teammates.

Compact Guard preserves critical context state before compaction and restores it after, so an agent can resume work without losing task progress, decisions, or next steps.