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/atopos31/llmio/agents-mdgit clone --depth 1 https://github.com/atopos31/llmioWrote 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/atopos31/llmio/agents-md)<a href="https://agentmods.dev/instructions/atopos31/llmio/agents-md"><img src="https://agentmods.dev/badge/instructions/atopos31/llmio/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 | $0.00937 | $0.00937 |
| Opus 5 | $0.00468 | $0.00468 |
| Sonnet 5 | $0.00187 | $0.00187 |
| Haiku 4.5 | $0.00094 | $0.00094 |
Grade A, and why
llmio AGENTS.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
Repository Guidelines
项目结构与模块组织
main.go负责启动 HTTP 服务与路由注册;核心业务拆分在service/handler/middleware/与providers/,对应业务服务、请求处理、拦截器和多模型供应商适配。common/汇聚通用工具,models/定义 GORM 实体并连接db/llmio.dbSQLite 数据库。- 前端代码位于
webui/,使用 React + TypeScript + Tailwind + Vite;构建产物输出至webui/dist/并由 Go 服务托管。 - 运维说明存放在
docs/,balancer/封装多供应商权重与请求调度策略。
构建、测试与开发命令
go run main.go本地启动 REST API 与管理后台,默认读取TOKEN等环境变量。make run依次执行go fmtgo mod tidygo run .,确保代码格式与依赖同步后再启动。go test ./...跑通所有 Go 单元测试,目前集中在handler/test_test.go,每次提交前需保持通过。- 使用
go test ./...若遇到权限不足,请在当前目录下设置GOCACHE=$PWD/.gocache后重试,测试结束立刻删除.gocache目录。 cd webui && pnpm install && pnpm run dev开启 Vite HMR ,pnpm run build生成生产静态资源。
代码风格与命名约定
- Go 代码统一使用
go fmt和 tab 缩进;导出符号使用 UpperCamelCase ,局部变量使用 lowerCamelCase ,JSON 标签使用 snake_case。 - 当前项目使用 Go 1.26;需要创建指针值时优先使用新版
new(...)语法(例如new(true)new(false)new(ioLog)),不要再引入boolPtr之类的辅助函数。 - 错误处理遵循
if err != nil模式,返回具备上下文的fmt.Errorf消息。 webui/src/中 React 组件采用 PascalCase 文件和组件名,目录建议 kebab-case;提交前执行pnpm run lint保持 ESLint + TypeScript 规范。
测试指南
- Go 测试文件与实现同目录,文件名追加
_test.go,函数命名Test<Feature>并优先采用表驱动验证路由、限流与供应商切换逻辑。 - 增量功能需补充针对中间件(鉴权、速率限制、流式响应)的覆盖;出现 bug 时先写回归测试再修复。
- 前端改动至少通过
pnpm run build或pnpm run preview做冒烟验证,并在 PR 中记录手动检查步骤。 - 如果可用
goplsMCP,每次代码编写完成后运行一次gopls检查,确保无诊断错误。
提交与 PR 指南
- 沿用 Conventional Commits 前缀,如
feat:fix:refactor:,必要时追加模块范围(例如feat(webui): adjust theme)。 - 每个提交聚焦单一问题,关联 Issue 或需求编号,可使用中文或英文摘要但需明确意图。
- PR 描述应包含背景、主要变更点、验证命令输出(
go testpnpm run build),前端改动附截图或录屏,并邀请相关维护者审阅。 - 在提交前确认未泄露 API Key、Token 等敏感信息,必要时使用
.env或 Secrets 管理。
安全与配置提示
- 供应商凭证放置于环境变量或未纳入版本控制的
.env文件,通过管理后台或迁移脚本写入providers表。 - 分享访问权限时及时轮换
TOKEN,生产部署通过反向代理时确认 TLS 终端配置正确,并对外暴露端口前启用访问控制。
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.
- 4d ago First seen · 37 lines · 937 tokens per session scan A 1ada268b73d0
llmio AGENTS.md is an instructions file published in the GitHub repository atopos31/llmio (312 stars, last pushed 18d ago), licensed MIT. It adds 937 tokens to every session, about $0.0047 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
phi AGENTS.md
Instructions for pulseaiclub/phi, covering phi, communication preferences, constraints, contributor guidelines and commands.
lightcode copilot-instructions.md
Instructions for Kartik-2239/lightcode, covering copilot instructions, build and test commands, high-level architecture and key conventions.
portage AGENTS.md
AGENTS.md instructions for SimonGino/portage, covering portage, 构建与测试, 与 po 协作, 事实源:查什么去哪 and 硬约束.
portage CLAUDE.md
Claude Code instructions for SimonGino/portage, a project described as: Keep the harness, change the model. A self-hosted single-binary gateway that lets Claude Code and Codex CLI run any model — open-weight, self-hosted, or from whatever relay you pay for — by translating between Anthropic Messages, OpenAI Responses…
ollama CLAUDE.md
Claude Code instructions for ollama/ollama: See AGENTS.md for the shared agent instructions for this repository.
gptme AGENTS.md
AGENTS.md instructions for gptme/gptme, covering agent instructions for gptme, git workflow, code style, testing and project structure.