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/briefgit 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 | $0.00012 | $0.01152 |
| Opus 5 | $0.00006 | $0.00576 |
| Sonnet 5 | $0.00002 | $0.00230 |
| Haiku 4.5 | $0.00001 | $0.00115 |
Grade A, and why
brief 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 3d 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 — 131 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/brief - Product Brief input contract
The Product Brief captures what to build, for whom, why, the MVP boundary, and known decisions. It is not a PRD, backlog, marketing document, research report, or task plan.
<input_contract>
Use /brief for a concept that is coherent enough to summarize, either directly
from operator input or after /brainstorm.
Read when present and relevant:
- the latest
.memory-bank/analysis/brainstorming/BR-*.md; .memory-bank/analysis/product-brief.md;.memory-bank/glossary.mdand the glossary row in.memory-bank/spec-index.md;.memory-bank/constitution.mdfor next-step routing;- operator-provided concept, notes, or PRD-like text.
If brainstorming exists, include it under Source artifacts. A report is not
required when the operator supplied a clear concept directly.
</input_contract>
<creator_vibe_contract>
Before interpreting source artifacts or interviewing the operator, load and
follow the installed creator-vibe skill. Apply it while reading the concept,
brainstorming report, notes, PRD-like text, and operator answers so the brief
preserves the intended human experience and what is distinctive about the
product.
Creator-vibe guides interpretation and phrasing only. It does not supply a missing decision, promote an inference to accepted product intent, or expand scope, value, MVP, success measures, constraints, or terminology without an explicit operator answer or authoritative source. </creator_vibe_contract>
<hard_invariants> Allowed outputs are only:
.memory-bank/analysis/index.md;.memory-bank/analysis/product-brief.md;.memory-bank/glossary.md;- the glossary row in
.memory-bank/spec-index.md.
Do not create features, task records, implementation plans, research reports,
PRFAQ, command aliases, or .memory-bank/analysis/assumptions.md. Preserve the
existing decision vocabulary proceed|blocked; do not add no-go.
</hard_invariants>
<operator_decisions> When a relevant ambiguity or branch can change product intent, users, scope, non-goals, value, constraints, success measures, or the PRD input, interview the operator before choosing. Questions are adaptive: ask one or group only tightly related decisions, use multiple choice or open form as useful, explain the impact, and optionally recommend an option.
A recommendation/default is not accepted until the operator explicitly answers.
Do not run a decorative questionnaire when authoritative evidence is already
clear. Put accepted decisions into the applicable brief sections and keep
unresolved material choices under Open Questions with Decision: blocked.
</operator_decisions>
<required_outputs>
Create or update .memory-bank/analysis/product-brief.md with required
frontmatter:
---
description: Product Brief input contract for PRD.
status: draft
type: product-brief
---
Keep the brief concise and cover:
- Metadata: status,
Decision: proceed|blocked, and source artifacts; - One-liner, Target Users, Problem, Current Alternatives;
- Value Proposition and Product Concept;
- MVP Scope and Non-goals;
- Success Metrics and Constraints;
- Assumptions, Risks, Open Questions;
- PRD Input Summary;
- final
## Decisionwithproceed|blocked.
If .memory-bank/analysis/index.md is created, give it analysis-index
frontmatter. Update the index with the brief status, source artifacts, decision,
and immediate next command.
After the brief content is formed, create or update
.memory-bank/glossary.md: remove bootstrap placeholders, preserve compatible
existing definitions, and add only terms grounded in the accepted brief. Do not
invent definitions or encode unresolved choices as facts. Keep the glossary row
in .memory-bank/spec-index.md consistent and include /brief in its change
route.
</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.
- 3d ago First seen · 131 lines · 12 tokens per session scan A f8259327153b
brief is a command published in the GitHub repository nicelight/DevRails-26 (22 stars, last pushed 5d ago), licensed MIT. It adds 12 tokens to every session and 1,152 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
git
Git operations with intelligent commit messages and workflow optimization.
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.
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.