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 instructions/alfredxw/denova/agents-mdgit clone --depth 1 https://github.com/alfredxw/denovaWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/alfredxw/denova/agents-md)<a href="https://agentmods.dev/instructions/alfredxw/denova/agents-md"><img src="https://agentmods.dev/badge/instructions/alfredxw/denova/agents-md.svg" alt="Measured on agentmods" height="20"></a>What 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 | $0.02541 | $0.02541 |
| Opus 5 | $0.01270 | $0.01270 |
| Sonnet 5 | $0.00508 | $0.00508 |
| Haiku 4.5 | $0.00254 | $0.00254 |
Grade A, and why
denova AGENTS.md 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 5d 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 — 90 lines — stays where its author put it; the contents beside it link to each section on GitHub.
项目目标
一个面向创作者的终极 AI 创作工具,支持完整创作流程,游戏交互
必须遵守
- 当前 Beta 的兼容基线仅为最近一个已发布版本的用户数据(当前为 v0.3.3);在简单可控范围内保护或迁移这些数据。内部接口、代码契约及最近一次 Release 后从未发布的中间格式不保留兼容,也不作为不兼容变更记录。
- 每次支持新的功能/功能较大变更时,需要考虑是否需要增加配置项供用户配置
- 模块依赖需要划分清晰,尽量避免单个文件/package负责太多不同的东西,原则上如果单文件功能太复杂那么就需要拆分成多个文件放到一个子package里。但文件和package也不能拆太碎,比如特别小的单文件package,可以合理范围内合并或者多层package划分
CHANGELOG.md的Unreleased以最近一个 Release 为比较基线,只记录升级用户最终可感知的重大功能、已发布用户数据的不兼容变化,以及已发布核心流程的重要修复;不记录内部接口、实现重构、测试调整、文案修改、细节级 UI 优化或未发布中间格式的变化与修复。条目保持中英文简短摘要,发布前再次压缩,并遵循 Keep a Changelog 规范。- 共通的能力支持/调整需要同时考虑到写作和游戏两个模式
- 发布版本时需要同步更新前端版本号、CHANGELOG.md,同步更新 README.md + README.en.md 内容,与项目功能保持一致,并创建对应 Git tag。发布 github release 时,Release notes 中需要简要说明该版本更新内容,特别是用户有明显感知的功能变更与修复,不需要把完整changelog复制过去,只需要brief。
- 能用依赖解决方案就不要自己实现,避免重复造轮子,多用组件库里的组件,比如 shadcn, tiptap, 在设计方案的时候,也优先考虑是否有成熟的开源组件可以使用。
- 系统与产品功能设计上多考虑功能是否真的有必要,新的功能是否可以被更高级的通用抽象支持,比如一般的提示词需求可以通过Skills实现,工具调用需求可以通过 Tools / Bash Execute 实现。
- 高度优先,追求极致。所有设计为长期服务,重点考虑设计是否优雅、边界是否清晰、是否长期可维护性高。
- 充分合理注释,涉及用户参与的设计决策的代码,需要明确清楚。
- 最终汇报时,用简单直白的语言说明:1 你做了什么。2 结果怎么样。3 你验证了什么。如果还有风险或限制,直接说清楚
- 除非用户指定,生产出的 docs (md/html) 默认放在 docs/ 目录下。
- 设计 agent tool 输入参数或者任何受限于 agent 输出的 schema 时,需要充分考虑到 agent 可能犯错的情况,尽可能去包容可修正的错误,以及对于list的情况下的部分成功与部分失败,避免大块输入整体失败导致昂贵的重试。
- 所有面向用户的交互,都要支持双语(展示中文和英文)
- 所有注入模型的提示词、上下文片段、工具描述、schema 描述和模型可见工具反馈统一使用英文,不得中英双语重复;仅面向用户的本地化交互保留中英双语。
- Denova 内置的模型可见内容保持产品中立,不得提及或类比其他 Agent 产品、CLI 或品牌;对照研究只保留在明确的审计文档中。
- 影响模型上下文相关设计的改动必须将缓存命中率(即前缀匹配)放在高优先级考虑事项。
- 所有注入模型的片段都要有明确来源、用途、需要注意硬上限需要配置高一些,避免随意截断。
- Do not preserve backward compatibility for internal APIs or unreleased formats. Protect user data from the latest release, using a small explicit migration only when necessary.
- Choose the simplest implementation that fully meets the current requirements. Avoid speculative abstractions, configuration, and indirection.
- Grow the system in layers. Start from the smallest version that works end to end, and add each new capability on top of a product that already works. Never trade a working product for unfinished complexity.
- Keep components modular and concerns clearly separated.
- Prefer established, well-maintained libraries when they reduce overall complexity or improve reliability. Do not reimplement common functionality without a clear reason.
- Lean on the dependencies already in the project before writing your own implementation or adding packages. Do not assume a library lacks a capability without checking its documentation and types.
- Make architectural decisions for the long term. Do not accept a stopgap that only works for now and is meant to be replaced later.
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.
- 5d ago First seen · 90 lines · 2,541 tokens per session scan A 771831f09d54
denova AGENTS.md is an instructions file published in the GitHub repository alfredxw/denova (673 stars, last pushed 7d ago), licensed Apache-2.0. It adds 2,541 tokens to every session, about $0.0127 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 instructions, from other repositories
neuro-book AGENTS.md
AGENTS.md instructions for notnotype/neuro-book, covering neurobook agent 入口, core rules, 了解开发者, 开发授权与通知 and 真实模型调用与样本数据.
neuro-book CLAUDE.md
Claude Code instructions for notnotype/neuro-book, a project described as: An AI-powered IDE for long-form fiction writing, combining software engineering workflows, modern storytelling methodologies, and multi-agent systems.
nanobrowser CLAUDE.md
Instructions for nanobrowser/nanobrowser, covering claude.md, project overview, development commands, workspace tips and architecture.
bifrost AGENTS.md
Instructions for bifrost-proxy/bifrost, covering bifrost 项目开发规则, 执行心法, 执行模式与完成定义, 任务模式判定 and 变更类型与验证路由(先判定,后执行).
octocode CLAUDE.md
Claude Code instructions for bgauryy/octocode: Read AGENTS.md. Discover repository skills under .agents/skills and activate only those matching the task.
Commonly-used-high-value-skills AGENTS.md
AGENTS.md instructions for seaworld008/Commonly-used-high-value-skills, covering repository agent instructions, architecture overview, installation root by client, golden rules and automated operations (sop for ai agents).