# Xiaohongshu Comment Reply

> Draft and diagnose natural Xiaohongshu comment replies, pinned comments, objection handling, discussion prompts, high-risk comment responses, and appropriate consultation invitations. Use for 小红书评论回复, 置顶评论, 评论区运营, 质疑处理, 批量回复, 人设语气, discussion continuation, or soft private-message guidance. Do not use it for a full content calendar or account diagnosis.

- **Type:** Skill
- **Install:** `agentstack add skill-shenyuannext-wepr-growth-skills-xiaohongshu-comment-reply`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [ShenyuanNext](https://agentstack.voostack.com/s/shenyuannext)
- **Installs:** 0
- **Category:** [Productivity](https://agentstack.voostack.com/c/productivity)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [ShenyuanNext](https://github.com/ShenyuanNext)
- **Source:** https://github.com/ShenyuanNext/wepr-growth-skills/tree/main/skills/xiaohongshu-comment-reply

## Install

```sh
agentstack add skill-shenyuannext-wepr-growth-skills-xiaohongshu-comment-reply
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# Operate Xiaohongshu Comments

Extend the conversation without sounding like a customer-service script or turning every reply into a sales action.

## Inputs

Use the post topic and body, original comment, account role and voice, known product or service facts, desired interaction, prohibited claims, and any sensitive context. When context is missing, default to calm, specific, human language and avoid a conversion push.

## Classify first

Classify each comment by motive:

- agreement or shared experience;
- question or request for detail;
- objection or skepticism;
- product/service fit inquiry;
- correction or factual dispute;
- complaint or unresolved service issue;
- provocation, abuse, spam, or safety risk.

Match the reply objective to the motive. Do not answer all comments with the same template.

## Reply method

1. Acknowledge the actual point or emotion without automatically agreeing with an incorrect claim.
2. Add one useful fact, boundary, question, or example.
3. Invite a next conversational step only when it is natural.
4. For objections, address the concern and evidence before mentioning the offer.
5. For complaints or legal/safety risk, stop public improvisation, preserve evidence, and route to the appropriate support, legal, or crisis workflow.
6. For abuse or spam, recommend hide/report/no reply when engagement adds risk.

## Voice rules

- Use short sentences and varied phrasing.
- Avoid repeated greetings, excessive exclamation marks, and copied customer-service closings.
- Do not be sarcastic, defensive, patronizing, or argumentative.
- Do not promise price, inventory, treatment, results, refunds, or private-message content unless confirmed.
- Do not move every interested user to private messages. Answer useful public questions publicly when appropriate.

## Pinned comments

Choose one job:

1. clarify a boundary or omitted detail;
2. add a save-worthy checklist;
3. invite specific experience-based discussion;
4. collect questions for a follow-up post;
5. explain a permitted next step without pressure.

## Output

For one comment, return a recommended reply, two tone alternatives, an optional next-step version, and a short risk note. For multiple comments, return a table with classification, recommended reply, alternative, conversion suitability, and escalation risk. For a pinned comment, provide three versions with distinct jobs.

If a comment is high risk, lead with the operational recommendation—reply, pause, hide, report, escalate, or document—before drafting any public text.

## Guardrails

Do not fabricate user details, transactions, promotions, scarcity, reviews, or resolutions. Do not encourage coordinated comments, sockpuppet replies, hidden advertising, harassment, doxxing, or moderation evasion. Do not diagnose legal, medical, or financial issues in comments.

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [ShenyuanNext](https://github.com/ShenyuanNext)
- **Source:** [ShenyuanNext/wepr-growth-skills](https://github.com/ShenyuanNext/wepr-growth-skills)
- **License:** MIT

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-shenyuannext-wepr-growth-skills-xiaohongshu-comment-reply
- Seller: https://agentstack.voostack.com/s/shenyuannext
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.