Oma Brainstorm
Design-first ideation: one question at a time, 2-3 labelled approaches, blind review
Test report
- Verdict
- Tested · Works
- Score
- Tested
- Jul 21, 2026
- Environment
- Claude Code 2.x (agent harness)
- Upstream re-checked
- Aug 10, 2026 · 2f4daf6
Found SKILL.md at .agents/skills/oma-brainstorm/SKILL.md (10,492 bytes); frontmatter has name+description, and all five referenced files returned HTTP 200 (resources/triz-lite.md, _shared/core/context-loading.md, clarification-protocol.md, reasoning-templates.md, skill-routing.md) — no hardcoded machine paths, no base64/eval/secret-exfil in the skill or the TRIZ resource, though the repo README's quick start is a `curl | bash` installer. Trigger phrasings judged: SHOULD fire — "I have an idea for a feature, let's brainstorm before we plan", "help me compare three approaches to rate limiting and pick one", "I want to design the notification system before we start building"; SHOULD NOT — "the offline sync is dropping updates, figure out why" (skill routes to oma-debug) and "break this approved spec into tickets with estimates" (oma-pm); all 5 decisions came out right, only caveat being name overlap with another installed brainstorm skill. Baseline vs skill on the same prompt: my unaided answer dumped four options as a menu with generic pros/cons and batched five questions at the END (the exact "jumping to solutions" pitfall the skill names), while the skill run asked one question first, produced three approaches with named scenario/mechanism/cost prose plus tactical-vs-structural labels before the matrix, and its mandatory blind-review round surfaced two Tier-1 defects the baseline never raised — an undefined delete-vs-edit race in the op log, and the full note corpus sitting unencrypted in IndexedDB. Docked points because the Actions/SSL-primitive tables and the `oma state:emit --checkpoint option-selection` step are OMA-runtime bookkeeping I could not run and that add nothing standalone, and the skill never warns that those bits need the `oma` CLI.
Scored on four weighted criteria — install, triggering, output vs. baseline, docs. How scoring works
- Installs cleanly 4/5
- Triggers reliably 5/5
- Output vs. baseline 7/10
- Docs & honesty 4/5
What Oma Brainstorm does
Runs an ideation workflow before any planning or coding: clarifying questions asked one at a time, then two or three mechanistically distinct approaches written as prose briefs (scenario, mechanism, what it solves and leaves, cost) and labelled tactical or structural, followed by a comparison matrix, a recommendation, and a blind-review round that critiques the design through 4-8 independent lenses into Tier 1/2/3 issues. The approved design is saved to docs/plans/designs/ before handoff to planning. Triggers on "I have an idea", "let's brainstorm", "compare these approaches", or "design this before we build it"; explicitly defers debugging, code review, and task decomposition to other skills.
How to install Oma Brainstorm
apm install first-fluke/oh-my-agent
# or manual (keeps the ../_shared/core references resolvable):
git clone --depth 1 https://github.com/first-fluke/oh-my-agent /tmp/oma
mkdir -p ~/.claude/skills
cp -R /tmp/oma/.agents/skills/oma-brainstorm ~/.claude/skills/
cp -R /tmp/oma/.agents/skills/_shared ~/.claude/skills/
# full framework (adds the `oma` CLI, hooks, workflows):
bunx oh-my-agent@latest
Skills live in ~/.claude/skills/ (global) or .claude/skills/
(per-project). Restart Claude Code after installing.
Commands — how to trigger Oma Brainstorm
-
/oma-brainstormDesign-first ideation: one question at a time, 2-3 labelled approaches, blind review
It also activates on plain-language prompts like these:
-
Brainstorm approaches for this new feature -
Explore concepts before we plan this out -
Evaluate different ways to solve this problem
Frequently asked questions
- Is the Oma Brainstorm skill free?
- Yes. The skill itself is free from first-fluke/oh-my-agent. SkillProof publishes the install command and an independent test verdict at no cost.
- Does Oma Brainstorm work with Claude Code?
- We tested it with Claude Code 2.x (agent harness) on Jul 21, 2026. Verdict: Tested · Works. Found SKILL.md at .agents/skills/oma-brainstorm/SKILL.md (10,492 bytes); frontmatter has name+description, and all five referenced files returned HTTP 200 (resources/triz-lite.md, _shared/core/context-loading.md, clarification-protocol.md, reasoning-templates.md, skill-routing.md) — no hardcoded machine paths, no base64/eval/secret-exfil in the skill or the TRIZ resource, though the repo README's quick start is a `curl | bash` installer. Trigger phrasings judged: SHOULD fire — "I have an idea for a feature, let's brainstorm before we plan", "help me compare three approaches to rate limiting and pick one", "I want to design the notification system before we start building"; SHOULD NOT — "the offline sync is dropping updates, figure out why" (skill routes to oma-debug) and "break this approved spec into tickets with estimates" (oma-pm); all 5 decisions came out right, only caveat being name overlap with another installed brainstorm skill. Baseline vs skill on the same prompt: my unaided answer dumped four options as a menu with generic pros/cons and batched five questions at the END (the exact "jumping to solutions" pitfall the skill names), while the skill run asked one question first, produced three approaches with named scenario/mechanism/cost prose plus tactical-vs-structural labels before the matrix, and its mandatory blind-review round surfaced two Tier-1 defects the baseline never raised — an undefined delete-vs-edit race in the op log, and the full note corpus sitting unencrypted in IndexedDB. Docked points because the Actions/SSL-primitive tables and the `oma state:emit --checkpoint option-selection` step are OMA-runtime bookkeeping I could not run and that add nothing standalone, and the skill never warns that those bits need the `oma` CLI.
- What is the Oma Brainstorm SkillProof Score?
- 8.0/10 — installs cleanly 4/5, triggers reliably 5/5, output vs. baseline 7/10, docs & honesty 4/5.
- How do I install Oma 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 Oma 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.