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/nicelight/devrails-26/brainstormgit clone --depth 1 https://github.com/nicelight/DevRails-26What 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 | $0.00017 | $0.01106 |
| Opus 5 | $0.00009 | $0.00553 |
| Sonnet 5 | $0.00003 | $0.00221 |
| Haiku 4.5 | $0.00002 | $0.00111 |
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 2d ago.
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 — 138 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/brainstorm - Facilitated ideation
The agent facilitates exploration and may recommend directions. The operator owns product intent, priorities, selections, and rejections.
<creator_vibe_intake>
At the start of /brainstorm, check for
.memory-bank/analysis/product-brief.md and existing
.memory-bank/analysis/brainstorming/BR-*.md reports. When both are absent,
load and follow the installed creator-vibe skill, then ask exactly one
standalone question and wait for the answer:
Опишите в нескольких предложениях самый важный или самый особенный функционал, который будет реализовывать ваше приложение.
Do not combine this intake with other questions. The answer may contain the
idea directly or point to an idea file. Before interpreting the answer, or
before reading the referenced idea file, load and apply creator-vibe again.
When either existing artifact is present, load and apply creator-vibe before
reading it, then continue from the unresolved decisions in that context. Carry
that lens through the adaptive interview and report without explaining the
interpretation back to the operator.
Creator-vibe may reveal intent, desired experience, and what must remain recognizably authored. It does not turn inferred intent into an accepted direction, requirement, priority, scope decision, or rejection. </creator_vibe_intake>
<input_contract>
Use /brainstorm when the problem space is unclear or several materially
different product directions are plausible. Read the operator's input and any
relevant existing analysis report or Product Brief before updating an existing
line of thought.
Create if missing:
.memory-bank/analysis/index.md.memory-bank/analysis/brainstorming/
Allocate the next free BR-<NNN>.md ID. If creating
.memory-bank/analysis/index.md, include frontmatter with
description: Analysis artifact index. and status: active.
</input_contract>
<hard_invariants>
- Raw ideas are not requirements or accepted product decisions.
- Do not create a PRD, Product Brief, feature/task records, implementation plan,
research report, PRFAQ, command alias, or separate
.memory-bank/analysis/assumptions.md. - Do not perform research in this command. Record research needs as risks or open questions.
- A recommendation is not an operator decision. Silence or continuation is not acceptance. </hard_invariants>
<operator_decisions> Interview adaptively whenever a real ambiguity or branch could change the selected direction, target users, scope, non-goals, constraints, or the Product Brief handoff. Ask one question or a small group of tightly related questions, using multiple choice or open form according to the decision. Explain what the choice changes and optionally recommend an option with rationale.
Do not ask questions merely to fill the report. When authoritative operator
input already resolves the issue, apply it without reconfirmation. Record only
explicitly accepted directions under Selected Directions; keep unresolved
choices under Open Questions and do not present them as decided.
</operator_decisions>
<required_outputs>
Create .memory-bank/analysis/brainstorming/BR-<NNN>.md with this minimum
shape:
---
description: Brainstorming report BR-<NNN>.
status: active
type: brainstorming-report
id: BR-<NNN>
---
# Brainstorming Report - BR-<NNN>
## Status
- Date: YYYY-MM-DD
- Facilitator: AI
- Source of product intent: operator
- Recommended next step: /brief
## Problem Space
## Goals
## Constraints
## Techniques Used
## Raw Ideas
## Clusters / Themes
## Selected Directions
## Rejected Directions
## Assumptions
## Open Questions
## Recommended Next Step
/brief
The id must match the filename. Assumptions and open questions stay in the
report. Update .memory-bank/analysis/index.md with the report link and its
immediate recommended next step.
</required_outputs>
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.
- 2d ago First seen · 138 lines · 17 tokens per session scan A 0dfd2c9406e8
brainstorm is a command published in the GitHub repository nicelight/DevRails-26 (22 stars, last pushed 4d ago), licensed MIT. It adds 17 tokens to every session and 1,106 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 2026-08-30.
Other commands, from other repositories
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.
constitution
Create or update the project constitution from interactive or provided principle inputs.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.