Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add commands/data-wise/craft/brainstormgit clone --depth 1 https://github.com/Data-Wise/craftWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00019 | $0.01403 |
| Opus 5 | $0.00010 | $0.00701 |
| Sonnet 5 | $0.00004 | $0.00281 |
| Haiku 4.5 | $0.00002 | $0.00140 |
Grade A, and why
brainstorm scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 114 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/brainstorm - ADHD-friendly Ideation
This command is a thin shim. The canonical behavior lives in the
brainstormskill (a sibling ofbrainstorm-insights, which now holds only the session-insights operation — see "Why this is a redesign" below). This file exists only to preserve the explicit/craft:brainstormslash entry point through the v2.34.0 → v3.0.0 migration.
--refine / --no-refine (prompt pre-processing)
Runs the prompt-refiner by default to pre-process your input. Pass --no-refine to skip.
Under --yes, the refiner auto-accepts Execute now (no confirm prompt).
Do NOT act on the raw argument. First invoke the prompt-refiner skill with
the argument and project context. Follow that skill's canonical flow
(before/after box → fenced refined-prompt block → 4-way confirm: Execute now /
Copy for elsewhere / Edit first / Skip; --yes or auto mode auto-accepts
Execute now). If the user picks "Copy for elsewhere," stop here — do not
start brainstorm's own depth/focus question flow. Otherwise proceed using
the prompt the skill returns. On no-argument interactive commands, refine
AFTER the topic is captured.
When invoked
- Load the canonical procedure: read
${CLAUDE_PLUGIN_ROOT:-.}/skills/workflow/brainstorm/SKILL.mdand follow it exactly — parse args, pick depth+focus, context scan, expert questions, one follow-up offer, generate output, optional spec capture, optional--orchhandoff. - Do not reimplement here. Any change to
/brainstormbehavior must be made in the skill, never duplicated into this shim.
Why this is a redesign, not just a shim
The pre-redesign command and the brainstorm-insights skill it pointed to
both carried full, diverging implementations of brainstorming — see the
deprecated-commands audit. This redesign:
- Splits brainstorm from insights.
brainstorm-insightsbundled two operations sharing no input or output: ideation (topic + context → BRAINSTORM/SPEC) and session-facet friction analysis (~/.claude/usage-data/facets/→ report). Insights now lives alone atskills/workflow/brainstorm-insights/(directory name kept for path stability — tests andcommands/workflow/insights.mdreference it). Brainstorm moved to its ownskills/workflow/brainstorm/. - Cut decision points from four to two. The old flow had separate depth and focus menus, a per-depth "ask more?" escape hatch with different options at each depth, and milestone re-prompts every N questions at high counts. Now: one combined depth+focus prompt, one follow-up offer after the expert questions.
- Removed in-skill agent delegation. The old "max" depth spawned 1-2
background subagents directly, using agent type names that didn't exist
(silently falling back to
general-purposewith no model pin — fixed earlier in this branch). That duplicated whatorchestrator-v2already does correctly (wave checkpoints, file-based results, explicit model routing, confirmation before spawning). Brainstorm no longer spawns agents itself — deep analysis needing multiple agents goes through--orch(routes toorchestrator-v2, unaffected by this redesign) after a spec is captured. One delegation mechanism, not two. There is no moremaxdepth; usedeepfor the full question set, then--orchif multi-agent implementation planning is wanted.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- yesterday First seen · 114 lines · 19 tokens per session scan A c0b0b16f2eae
brainstorm is a command published in the GitHub repository Data-Wise/craft (4 stars, last pushed 16d ago), licensed MIT. It adds 19 tokens to every session and 1,403 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other commands, from other repositories
readme-audit
Audit a README's reading flow by treating it as a conversion funnel. Every section is evaluated against a target reader and a terminal action. Produces a structured audit blueprint that feeds directly into /readme-restructure.
readme-restructure
Execute a /readme-audit blueprint. Rewrites the README to the evaluator → conversion → links structure and creates linked docs by extracting and reorganising existing content.
build-fix
Fix build and type failures with minimal diffs.
checkpoint
Record a verified checkpoint before the next phase.
doctor
Check the install surface for missing files and invalid manifests.
feature-dev
Drive a feature from plan to implementation to review.