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 agents/xerrors/yuxi/middlewaregit clone --depth 1 https://github.com/xerrors/YuxiWhat 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 | $0.00000 | $0.01459 |
| Opus 5 | $0.00000 | $0.00730 |
| Sonnet 5 | $0.00000 | $0.00292 |
| Haiku 4.5 | $0.00000 | $0.00146 |
Grade A, and why
middleware 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 — 77 lines — stays where its author put it; the contents beside it link to each section on GitHub.
中间件
中间件把文件、Skills、子智能体、上下文压缩、审批和用量统计接到 LangGraph Agent。它们在模型调用、工具调用或 state 更新的边界运行，让不同 Agent 复用同一套能力。
内置 ChatbotAgent 和 SubAgentBackend 都在 get_graph() 中组装中间件。Graph 创建前，系统先完成用户资源和权限的归一化；中间件不应绕过这一步重新决定授权。
Graph 创建前的准备
prepare_agent_runtime_context 会根据当前用户和 Agent 配置：
- 过滤内置工具、知识库、MCP、Skills 和子智能体；
- 生成
_visible_knowledge_bases； - 展开 Skill 依赖，生成
_effective_skill_slugs和_runtime_skills； - 使用系统默认模型补齐空的模型配置。
随后，工具解析器准备可执行工具，build_prompt_with_context 生成系统提示词，Agent 再创建 Graph。工具执行时仍需检查具体目标，准备阶段的资源快照不是授权替代品。
内置中间件顺序
ChatbotAgent 的常见顺序如下；可选项只在对应能力启用时加入：
| 顺序 | 中间件 | 作用 |
|---|---|---|
| 1 | SteerMiddleware |
在安全边界发现待接替请求 |
| 2 | create_agent_filesystem_middleware |
提供 Workdir、User Data、Skills 文件后端，并卸载过大的工具结果 |
| 3 | SkillsMiddleware |
注入 Skill 说明，按激活状态开放依赖 |
| 4 | YuxiMemoryMiddleware |
Memory 开关开启且 MEMORY.md 有内容时，注入用户记忆并提供受限工具 |
| 5 | YuxiSubAgentMiddleware |
主智能体有可见子智能体时提供 task 和生命周期工具 |
| 6 | YuxiSummarizationMiddleware |
在上下文过大时执行 L1/L2 压缩 |
| 7 | TodoListMiddleware |
保存待办，供状态面板展示 |
| 8 | PatchToolCallsMiddleware |
修正部分工具调用消息形态 |
| 9 | ModelRetryMiddleware |
按配置重试模型调用失败 |
| 10 | ImageInputCompatibilityMiddleware |
桥接工具读取图片与模型输入格式；必要时回退 OCR |
| 11 | TokenUsageMiddleware |
记录近似上下文和主模型实际用量 |
| 12 | 工具审批 middleware | 默认模式下拦截写文件、编辑文件和执行命令 |
SubAgentBackend 复用文件、Skills、Summary、待办、重试和用量等能力，但不挂载子智能体 middleware，并过滤不适合子智能体的敏感或交互工具。
Skills 和知识库
Skills middleware 将 Skill 说明按模型请求注入：预加载 Skill 从首轮开放依赖，普通 Skill 在模型读取对应 SKILL.md 后激活，再开放声明的工具和 MCP。
知识库能力由内置 knowledge-base Skill 提供。它的工具是否注册、模型是否可见、参数是否能访问目标知识库分别由工具组装、Skill 激活和知识库权限检查负责。完整链路见工具系统和知识库机制详解。
文件和附件
附件确认后写入当前 Project Workdir。每次 Run 会把线程历史附件的文件名和实时路径加入本轮用户消息，让模型按需调用 read_file；持久化 Message 仍保存原始文本，不会把这段模型专用路径混进用户可见消息。
普通 Agent 和子 Agent 使用根 Conversation 的同一个 runtime_scope_id 和 Workdir。子 Agent 的 child thread 只隔离 LangGraph checkpoint，不隔离共享文件。Viewer、附件和 artifact API 直接访问 UserWorkspace 的持久文件，不需要创建 file-bridge Sandbox。
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 · 77 lines · 0 tokens per session scan A 477950f7bb9c
middleware is an agent published in the GitHub repository xerrors/Yuxi (6,591 stars, last pushed 2d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,459 tokens. 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 agents, from other repositories
txtify-qa
QA gatekeeper for Txtify. Use to run the full verification ladder on the current tree and report evidence — before merging, releasing, or when asked "does everything still work?".
agents
Agent 模块是系统的执行引擎，基于 LangGraph StateGraph 实现单个 Agent 的完整执行生命周期管理。.
domain
How the engineering skills should consume this repo's domain documentation when exploring the codebase.
2026-04-20-relay-handoff
Demo Agent Ada found that Relay parser failures were caused by missing fenced-code terminators in synthetic notes.
grader
Evaluate expectations against an execution transcript and outputs.
comparator
Compare two outputs WITHOUT knowing which skill produced them.