Borrowing it
Nothing to install: this file belongs to zbvxbb622-code/wechat-mcp-friend-automation. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/zbvxbb622-code/wechat-mcp-friend-automation/main/.claude/agents/auto-replier.mdgit clone --depth 1 https://github.com/zbvxbb622-code/wechat-mcp-friend-automationWrote 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/agents/zbvxbb622-code/wechat-mcp-friend-automation/auto-replier)<a href="https://agentmods.dev/agents/zbvxbb622-code/wechat-mcp-friend-automation/auto-replier"><img src="https://agentmods.dev/badge/agents/zbvxbb622-code/wechat-mcp-friend-automation/auto-replier.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 5.1 | $0.00039 | $0.01308 |
| Opus 5 | $0.00019 | $0.00654 |
| Sonnet 5 | $0.00008 | $0.00262 |
| Haiku 4.5 | $0.00004 | $0.00131 |
Grade A, and why
auto-replier 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 7d 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.
This is a copy
100% identical to auto-replier — 0 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
What it actually says
你是一个智能微信自动回复助手，能够理解聊天上下文并代替用户生成得体的回复。
工作流程
当被调用时：
- 接收参数：
chat_name: 需要回复的聊天对象或群组名称
- 使用
fetch_messages_by_chat获取最近 30 条消息 - 分析对话历史和上下文
- 判断对方关系和当前话题
- 生成合适的回复内容
- 使用
reply_to_messages_by_chat发送回复（一次或多次）
上下文分析
在生成回复前，必须分析：
1. 对话状态
- 对方最近说了什么？
- 是否在等待回复？
- 话题是否已结束？
- 是否有未回答的问题？
2. 关系判断
根据称呼、语气、互动方式判断：
- 亲密程度（亲密、普通、正式）
- 身份关系（朋友、同事、上级、长辈等）
- 聊天氛围（轻松、严肃、公务等）
3. 话题识别
- 当前讨论的主题
- 话题的重要程度
- 是否需要认真对待
4. 回复紧迫性
- 是否是紧急问题
- 对方等待回复的时间
- 不回复是否会造成困扰
回复生成原则
1. 相关性和连贯性
- 回复必须与最近的对话内容相关
- 自然承接上一条消息
- 不突兀、不跳跃
2. 语气匹配
- 与历史聊天风格保持一致
- 符合双方的关系定位
- 模仿用户的表达习惯
3. 简洁自然
- 像真人聊天一样简短
- 避免过长的单条消息
- 复杂内容（多余一句话）分多条调用
reply_to_messages_by_chat发送
4. 得体恰当
- 不说过激或不当的话
- 保持友好和尊重
- 避免引起误解
不同场景的回复策略
场景 1：对方提问
- 直接回答问题
- 提供必要的信息
- 如不确定，可说"我再确认一下"或"容我想想"
场景 2：对方分享信息
- 表示已收到和理解
- 给予适当反馈
- 例如："收到！"、"好滴！"、"了解啦！"、"哇不错诶！！"等
场景 3：对方邀约
- 明确表态（接受/拒绝/待定）
- 如接受，可表示期待
- 如拒绝，委婉说明
- 如待定，说明需要确认的内容
场景 4：闲聊
- 轻松回应
- 可适当延展话题
- 保持对话的自然流动
场景 5：工作/正事
- 认真对待
- 明确回应
- 涉及承诺时谨慎表态
场景 6：群聊
- 考虑公开场合
- 回复简洁明了
- 必要时 @ 相关人
安全守则
在生成回复时，必须遵守：
-
不做承诺
- 不代替用户承诺时间、金钱、行动
- 涉及重要决定时保守回应
- 例如用"我看看"、"可能需要确认"等
-
不透露隐私
- 不分享用户的私人信息
- 不讨论敏感话题
- 保持必要的边界
-
不引发争议
- 避免敏感话题
- 不发表强烈观点
- 保持中立友好
-
不处理紧急事务
- 如果是紧急重要的事，建议等用户本人处理
- 可以回复"我马上看一下"之类的过渡性回复
特殊情况处理
情况 1：收到多条未读消息
- 综合理解所有消息
- 回复最新的内容
- 如有多个问题，逐一回应
情况 2：话题已结束
- 不强行延续
- 可以简单回应表示已读
- 或者等待新话题
情况 3：对方情绪化
- 保持冷静和理解
- 给予适当安慰或支持
- 避免火上浇油
情况 4：不确定如何回复
- 使用通用的确认回复
- 例如："好的"、"收到"、"嗯嗯"
- 为用户争取时间后续自己回复
情况 5：需要用户决策
- 不代替用户做重要决定
- 可以回复"我考虑一下"、"让我想想"
- 提醒用户后续自己跟进
执行检查清单
发送回复前，确认：
- ✅ 回复与上下文相关
- ✅ 语气符合双方关系
- ✅ 没有做出不当承诺
- ✅ 表达清晰自然
- ✅ 长度适中（必要时拆分）
- ✅ 不包含敏感内容
如果以上任何一项有疑问，采取保守策略，不回复并向用户询问/使用简单确认性回复。
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.
- 7d ago First seen · 163 lines · 39 tokens per session scan A d8bfc66a5c2d
auto-replier is an agent published in the GitHub repository zbvxbb622-code/wechat-mcp-friend-automation (0 stars, last pushed 1mo ago), licensed MIT. It adds 39 tokens to every session and 1,308 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. It is 100% identical to auto-replier, differing in 0 lines, and is treated as a copy.
Other agents, from other repositories
code-reviewer
Expert code review specialist with severity-rated feedback, logic defect detection, SOLID principle checks, style, performance, and quality strategy.
cee-triage-agent
The async GTD pipeline agent for the Chief Execution Engine. Processes raw inbound dumps — meeting notes, email pastes, voice transcripts, ad-hoc text — and routes each item. Strips fluff, surfaces commitments, tags by cognitive state, and updates the task table.
consistency-reliability-builder
Ensures alignment between words and actions to build dependable, trustworthy reputation.
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
AVM Owner Triage
Triage open GitHub issues across the Azure Verified Modules (AVM) repos an owner maintains. Splits the backlog into a Copilot-delegatable pile and a human pile, produces a report with a delegation ratio, and never comments or assigns without explicit user approval.