prompt
FreeParallel divergent ideation for coding agents
FreeFree tier
About prompt
This is a prompt/skill designed for AI coding agents to perform parallel divergent ideation under multiple cognitive frames. It spawns N isolated branches representing different perspectives (e.g., regulator, biology, speedrunner, 10-year-old, $0 budget), then scores, clusters, prunes traps, and deepens the top survivors. The skill includes a pre-flight gating system to determine whether the expensive exploration is warranted—it is intended for open-ended, high-stakes problems such as architecture decisions, public API surfaces, naming, and fuzzy debugging. It is MIT licensed and hosted on GitHub as part of the ai-boost/awesome-prompts repository.
Key Features
Spawns isolated branches under different cognitive frames (regulator, biology, speedrunner, 10-year-old, $0 budget)
Scores, clusters, and prunes traps from generated ideas
Deepens top survivors for further exploration
Includes a pre-flight gate that checks if the problem is open-ended, high-stakes, and not explicitly requesting a direct answer
Explicit invocation via /adhd or ADHD mode bypasses the gate
Pros & Cons
Pros
- Encourages non-obvious, innovative solutions beyond the first few textbook answers
- Structured exploration with multiple cognitive perspectives
- Explicit trap detection and pruning reduces wasted effort
- Pre-flight gate prevents misuse on simple queries
Cons
- Expensive in terms of API calls (5-10x a single answer) and time (30-90 seconds wall clock)
- Not suitable for straightforward, canonical, or low-stakes queries
- Requires an AI agent that can execute multiple branches in parallel
Best For
Brainstorming and ideation for coding problemsOpen-ended design and architecture decisionsNaming products or featuresDesigning API/SDK surfacesFuzzy debugging with no known root cause
FAQ
What is ADHD parallel ideation?
It is a prompt/skill that instructs an AI coding agent to spawn multiple isolated reasoning branches under different cognitive frames, score and cluster ideas, prune traps, and deepen the best survivors to produce more creative and diverse solutions.
When should I use this skill?
Use it for open-ended, high-stakes problems like architecture decisions, public API surfaces, naming, and fuzzy bugs with no known root cause. Avoid it for syntax lookups, quick fixes, or when a canonical answer exists.
How do I invoke the skill?
You can type /adhd or explicitly request ADHD mode. Otherwise, the pre-flight gate will self-judge whether the problem qualifies for this expensive exploration.
Is this skill free?
The prompt itself is open-source (MIT license) and free to use. However, running the skill requires API calls to an AI model, which may incur costs depending on the provider.