Core Brainstorm
Socratic brainstorming mode that interrogates ideas and outputs a structured brief
Test report
- Verdict
- Tested · Works
- Score
- Tested
- Jul 31, 2026
- Environment
- Claude Code 2.x (agent harness)
- Upstream re-checked
- Aug 10, 2026 · 13f4170
Fetched SKILL.md via raw GitHub; frontmatter parses with name+description, plus argument-hint and disable-model-invocation:true. Body references no scripts/bundled files (only runtime output paths under .claude/brainstorms/), so nothing to 404; clean of security smells. Cloned into a temp HOME and confirmed SKILL.md lands at ~/.claude/skills/core-brainstorm/SKILL.md. Ran one real task ("should I add a public API to my solo-dev SaaS?"): baseline gave a one-shot Pros/Cons + "How to build it" answer naming 7 implementation terms (REST, Redis, OpenAPI, Swagger, /v1/, OAuth, token bucket); the skill-followed run contained 0 of those terms, instead running a multi-turn Socratic pushback that reframed "build an API" into "close two leads with read-only export" and closed with the 6-section Brainstorm Brief the baseline never produced. Trigger description cleanly separates brainstorm-intent from code/fix requests (5/5 on my test phrasings); the disable-model-invocation flag means it fires as an explicit /command, not an auto-load.
Scored on four weighted criteria — install, triggering, output vs. baseline, docs. How scoring works
- Installs cleanly 5/5
- Triggers reliably 5/5
- Output vs. baseline 8/10
- Docs & honesty 5/5
What Core Brainstorm does
A collaborative brainstorming mode that pushes back on assumptions, keeps responses short and question-led, refuses code/tool/implementation discussion, and ends by distilling the conversation into a structured "Brainstorm Brief" (problem, core idea, decisions, open questions, constraints) meant to hand off to a planner. Triggers when the user wants to discuss, explore, or think through a problem or feature before writing code. Ships with disable-model-invocation set, so it is invoked explicitly as /core-brainstorm rather than auto-loaded.
How to install Core Brainstorm
git clone --depth 1 https://github.com/LukeRenton/explore-claude-code.git /tmp/core-brainstorm-src
mkdir -p ~/.claude/skills
cp -R /tmp/core-brainstorm-src/.claude/skills/core-brainstorm ~/.claude/skills/core-brainstorm
rm -rf /tmp/core-brainstorm-src
# No dependencies, no scripts — single self-contained SKILL.md.
# Note: frontmatter sets disable-model-invocation: true, so Claude will NOT
# auto-load it. Invoke it explicitly, e.g. /core-brainstorm <topic>
# It is designed as step 1 of a brainstorm -> core-planner -> implement workflow;
# core-planner is a separate agent in the same repo and is not required to use this.
Skills live in ~/.claude/skills/ (global) or .claude/skills/
(per-project). Restart Claude Code after installing.
Commands — how to trigger Core Brainstorm
-
/core-brainstormSocratic brainstorming mode that interrogates ideas and outputs a structured brief
It also activates on plain-language prompts like these:
-
Let's brainstorm approaches to this new feature -
Think through this problem with me before planning -
Explore different ideas for the onboarding flow
Frequently asked questions
- Is the Core Brainstorm skill free?
- Yes. The skill itself is free from LukeRenton/explore-claude-code. SkillProof publishes the install command and an independent test verdict at no cost.
- Does Core Brainstorm work with Claude Code?
- We tested it with Claude Code 2.x (agent harness) on Jul 31, 2026. Verdict: Tested · Works. Fetched SKILL.md via raw GitHub; frontmatter parses with name+description, plus argument-hint and disable-model-invocation:true. Body references no scripts/bundled files (only runtime output paths under .claude/brainstorms/), so nothing to 404; clean of security smells. Cloned into a temp HOME and confirmed SKILL.md lands at ~/.claude/skills/core-brainstorm/SKILL.md. Ran one real task ("should I add a public API to my solo-dev SaaS?"): baseline gave a one-shot Pros/Cons + "How to build it" answer naming 7 implementation terms (REST, Redis, OpenAPI, Swagger, /v1/, OAuth, token bucket); the skill-followed run contained 0 of those terms, instead running a multi-turn Socratic pushback that reframed "build an API" into "close two leads with read-only export" and closed with the 6-section Brainstorm Brief the baseline never produced. Trigger description cleanly separates brainstorm-intent from code/fix requests (5/5 on my test phrasings); the disable-model-invocation flag means it fires as an explicit /command, not an auto-load.
- What is the Core Brainstorm SkillProof Score?
- 9.2/10 — installs cleanly 5/5, triggers reliably 5/5, output vs. baseline 8/10, docs & honesty 5/5.
- How do I install Core Brainstorm?
- Copy the install command from this page, run it in your terminal, and restart Claude Code. Skills live in ~/.claude/skills/ (global) or .claude/skills/ inside a project.
- Can I use Core Brainstorm with Cursor, Copilot, Gemini CLI, Codex or other AI tools?
- The SKILL.md format is native to Claude (Claude Code, Desktop, claude.ai). The instructions inside adapt to other assistants: Cursor rules, GitHub Copilot instructions, Windsurf rules, Custom GPTs, AGENTS.md for OpenAI Codex, and GEMINI.md for Google Gemini CLI — our conversion guides cover each, and the free converter on the tools page does the wrapping for you.