← Back to Skills Marketplace
pippin1214

RedNote Research

by Pippin · GitHub ↗ · v0.4.0 · MIT-0
cross-platform ⚠ suspicious
304
Downloads
0
Stars
1
Active Installs
5
Versions
Install in OpenClaw
/install rednote-research
Description
Research a topic through RedNote/Xiaohongshu discussion signals using either public-web mode (no login) or optional login-enhanced browser review when the us...
Usage Guidance
This skill appears coherent and safe for research tasks: it defaults to public-web searches and only uses a login-enhanced browser path if the user explicitly chooses to log in in a controlled session. Before installing, confirm your agent platform supports interactive/browser automation (the skill expects to open a browser session for user login). Never enter passwords into chat — follow the skill's instructed browser-native login flow. If you plan to run any included scripts, you can inspect them locally first (they are simple, readable utilities) and ensure your runtime won't unexpectedly execute external downloads.
Capability Analysis
Type: OpenClaw Skill Name: rednote-research Version: 0.4.0 The rednote-research skill bundle provides comprehensive tools for OSINT on the RedNote (Xiaohongshu) platform, including query expansion and claim normalization. It is classified as suspicious because it includes high-risk capabilities and potential vulnerabilities: the 'login-enhanced' mode involves browser automation and authenticated sessions, and the `scripts/claim_log_tools.py` utility performs unvalidated file reads via `Path(path).read_text()`, which could be exploited for arbitrary file disclosure if the agent is manipulated. While the documentation (SKILL.md and references) includes defensive instructions and safety guidelines for handling user logins, the combination of session-based browser access and lack of input sanitization in file-handling scripts meets the threshold for risky behavior without clear malicious intent.
Capability Assessment
Purpose & Capability
The name/description (RedNote/Xiaohongshu research) matches the included materials: detailed SKILL.md, query builders, claim-log tools, and many reference templates for evidence handling. There are no unrelated required env vars, binaries, or installs that don't belong to a web-research/analysis skill.
Instruction Scope
Instructions are specific and constrained to public-web searching and an explicit, user-consented 'login-enhanced' browser review. The SKILL.md explicitly forbids asking for passwords in chat and requires the user to complete the login inside the browser session. Note: the skill assumes an agent/platform-provided browser automation environment for the login-enhanced flow — that capability is referenced but not declared as a required binary in metadata (this is a reasonable assumption on many agent platforms, but you should confirm the runtime supports interactive browser sessions).
Install Mechanism
No install spec is provided (instruction-only install), so nothing is downloaded or extracted. The repository includes small utility scripts (query builder, recovery builder, claim-log tools) which are local and readable; they don't perform network calls or hidden downloads in the provided source.
Credentials
The skill requests no environment variables, no credentials, and no config paths. Its optional login flow is explicitly user-driven and does not ask for credentials via chat, which is proportionate for the stated purpose.
Persistence & Privilege
The skill is not marked always:true and has default autonomous invocation allowed (normal for skills). It does not request permanent presence or elevated system privileges, nor does it attempt to modify other skills or system configuration.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install rednote-research
  3. After installation, invoke the skill by name or use /rednote-research
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v0.4.0
Make the skill operational for real use: add a concrete login-enhanced workflow, account-summary templates, minimal no-login user-input fallback paths, and stronger guidance for hybrid public-web plus user-provided evidence research.
v0.3.0
Add dual access modes to the skill: default public-web research plus optional login-enhanced browser review when the user explicitly chooses deeper coverage. Document the login decision path, account-summary escalation flow, and clearer caveats for partial no-login coverage.
v0.2.0
Expand the RedNote/Xiaohongshu public-web research skill with multimodal analysis guidance, recovery-oriented query generation from snippets/OCR/subtitles, structured claim-log tooling, stronger verification patterns, a scoring rubric, and reusable report templates.
v0.1.1
Privacy scrub; stronger workflows and reliability caveats; improved query builder with aliases/geo/verification families; added verification patterns reference.
v0.1.0
Initial release: public-web RedNote/Xiaohongshu reputation research workflow with query builder and scoring rubric.
Metadata
Slug rednote-research
Version 0.4.0
License MIT-0
All-time Installs 1
Active Installs 1
Total Versions 5
Frequently Asked Questions

What is RedNote Research?

Research a topic through RedNote/Xiaohongshu discussion signals using either public-web mode (no login) or optional login-enhanced browser review when the us... It is an AI Agent Skill for Claude Code / OpenClaw, with 304 downloads so far.

How do I install RedNote Research?

Run "/install rednote-research" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is RedNote Research free?

Yes, RedNote Research is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does RedNote Research support?

RedNote Research is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created RedNote Research?

It is built and maintained by Pippin (@pippin1214); the current version is v0.4.0.

💬 Comments