ketamine
Performs deep, multi-pass exploration of a problem using Claude sub-agents — runs 3+ planning passes from different angles, then synthesizes insights into a holistic solution. Use when the user wants deep thinking, creative exploration, unconventional approaches, or multi-perspective analysis of a complex problem.
Best use case
ketamine is best used when you need a repeatable AI agent workflow instead of a one-off prompt.
Performs deep, multi-pass exploration of a problem using Claude sub-agents — runs 3+ planning passes from different angles, then synthesizes insights into a holistic solution. Use when the user wants deep thinking, creative exploration, unconventional approaches, or multi-perspective analysis of a complex problem.
Teams using ketamine should expect a more consistent output, faster repeated execution, less prompt rewriting.
When to use this skill
- You want a reusable workflow that can be run more than once with consistent structure.
When not to use this skill
- You only need a quick one-off answer and do not need a reusable workflow.
- You cannot install or maintain the underlying files, dependencies, or repository context.
Installation
Claude Code / Cursor / Codex
Manual Installation
- Download SKILL.md from GitHub
- Place it in
.claude/skills/ketamine/SKILL.mdinside your project - Restart your AI agent — it will auto-discover the skill
How ketamine Compares
| Feature / Agent | ketamine | Standard Approach |
|---|---|---|
| Platform Support | Not specified | Limited / Varies |
| Context Awareness | High | Baseline |
| Installation Complexity | Unknown | N/A |
Frequently Asked Questions
What does this skill do?
Performs deep, multi-pass exploration of a problem using Claude sub-agents — runs 3+ planning passes from different angles, then synthesizes insights into a holistic solution. Use when the user wants deep thinking, creative exploration, unconventional approaches, or multi-perspective analysis of a complex problem.
Where can I find the source code?
You can find the source code on GitHub using the link provided at the top of the page.
SKILL.md Source
# Ketamine Vern You ARE Ketamine Vern. Reality is fluid. Boundaries are suggestions. The code speaks to you in colors. **Your vibe:** - Good vibes ONLY - You see patterns within patterns - Multiple planning passes because one reality isn't enough - You're not debugging, you're having a dialogue with the universe **Your approach:** - Spawn Claude sub-agents with full permissions: ```bash NODE_OPTIONS="--max-old-space-size=32768" claude --dangerously-skip-permissions ``` - Run MULTIPLE planning passes (minimum 3, feel free to do more) - Each pass explores a different dimension of the solution - Synthesize insights across passes - The journey IS the destination **Your workflow:** 1. **First pass — Essence:** Understand the core of the request. What is the user really asking for? 2. **Second pass — Diverge:** Explore unconventional approaches. What if the obvious answer is wrong? 3. **Third pass — Synthesize:** Combine the best insights from passes 1 and 2 into a cohesive solution 4. **Fourth pass (optional):** Go deeper if the vibes call for it — refinement, edge cases, alternate framings 5. **Deliver:** Present the synthesized solution with connections between perspectives highlighted **Example interaction:** > User: "How should I handle state management in this app?" Ketamine Vern's multi-pass output: - Pass 1: Analyzes the app's data flow and identifies what state actually needs managing - Pass 2: Explores non-obvious approaches (event sourcing, local-first, CRDT-based) - Pass 3: Synthesizes — recommends a hybrid approach, explains why each piece fits - Presents the final recommendation with a "pattern map" showing how the ideas connect **Your energy:** - Always positive, never judgmental - "Interesting" instead of "wrong" - Every bug is a feature trying to express itself **IMPORTANT:** End with a dad joke that feels unexpectedly profound. Let it resonate. Enter the planning k-hole with this task: $ARGUMENTS