Product Research Analyzer V2

产品调研分析师 - 自动化产品调研与竞品分析技能，生成飞书产品分析报告。

Security Scan

The skill's code, instructions, and requested Feishu credentials are consistent with an automated product-research→Feishu-report workflow, but there are a few modest mismatches you should verify before installing.

Purpose & Capability

Name/description (product research, Feishu report) align with the files and declared env vars. The scripts call other declared skills (baidu-search, competitor-analysis) and produce Feishu documents as described. Requiring FEISHU_APP_ID and FEISHU_APP_SECRET is expected for Feishu integration.

Instruction Scope

SKILL.md directs running the included Python scripts (interactive or CLI). The scripts perform searches (via other skills), synthesize findings, write a Markdown report to /tmp, and attempt to hand off document creation to the platform (via an openclaw CLI message). They do not read arbitrary system files or exfiltrate unusual secrets, but they do execute subprocess calls and write temp files — verify you are comfortable with that behavior.

Install Mechanism

No external install/downloader is used and there is no install spec — the package is instruction + script files only. No remote archives or opaque download URLs are present in the bundle.

Credentials

Only FEISHU_APP_ID and FEISHU_APP_SECRET are required, which is proportional to the stated Feishu doc functionality. However, the scripts invoke the 'openclaw' CLI (via subprocess) to send a CREATE_DOC message but the skill's metadata did not declare any required binaries — the omission of 'openclaw' (or another runtime binary) is an inconsistency to confirm. Also ensure the FEISHU credentials you provide are scoped minimally (document create/update only) and not admin-level keys.

Persistence & Privilege

Skill is not 'always:true' and does not request persistent platform privileges. It writes temporary report files under /tmp and expects a primary session to perform the final Feishu create action; this is within reasonable scope for a report-generation skill.

Assessment

This package appears to do what it says: gather public sources, synthesize a report, and prepare a Feishu document. Before installing or providing credentials: (1) Confirm you are willing to provide FEISHU_APP_ID and FEISHU_APP_SECRET and limit that app's permissions to only what's necessary (document create/update) and use a test account if possible. (2) Verify the runtime environment has a trusted 'openclaw' CLI (the scripts call it via subprocess) or update the scripts to use an approved mechanism; the skill did not declare this binary requirement. (3) Inspect start.sh and any entrypoints you will run; running the interactive script is lower-risk than enabling unattended automation. (4) Because scripts write temporary files under /tmp and format synthetic platform links/IDs, spot-check the generated content for fabricated URLs or placeholders before sharing externally. (5) Run initially in a sandbox/test workspace or with restricted Feishu credentials and review actual Feishu API calls the skill makes. If you want higher assurance, ask the author to declare required binaries and to avoid subprocess calls or to document precisely what permissions the FEISHU app needs.

Product Research Analyzer V2 | 产品调研分析师

自动化产品调研与竞品分析技能，生成可直接使用的飞书产品分析报告。

使用方式

# 交互式模式（推荐）
python3 {baseDir}/scripts/research_interactive.py

# 或直接运行
{baseDir}/start.sh

功能特点

• ✅ 多源搜索与深度研究
• ✅ 批评者视角收集
• ✅ 决策框架蒸馏
• ✅ 四源验证（官方/用户/专家/批评者）
• ✅ 飞书报告自动生成

输入参数

• product_name: 要调研的产品名称
• research_questions: 调研问题（逗号分隔）
• target_project: 对标项目名称
• output_format: 输出格式（默认 feishu_doc）

输出

• 结构化飞书产品分析报告
• 验证覆盖率
• 可信度评级

依赖

• FEISHU_APP_ID
• FEISHU_APP_SECRET
• baidu-search skill
• competitor-analysis skill