Kol ClawSAFE
把 KOL 投放判断力接入任意 AI:达人评级、CPM 定价、话术生成,6 个 MCP 工具,直接给结论
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
把 KOL 投放的核心判断力,接入任意 AI。
达人值不值投、报价贵不贵、建联说什么 —— 这些决策以前靠经验靠感觉,现在封装成 6 个 MCP 工具,Claude Code 或任何支持 MCP 协议的 AI 都能直接调用,实时给出结论。
配合 kol-claw 主系统,实现从「一句话需求」到「完成建联」的全链路自动化。
能做什么
注册这个 MCP 服务之后,你可以直接对 AI 说:
分析一下这个达人:粉丝 8 万,最近5条播放 3万/2.8万/3.5万/2.5万/4万,报价 ¥2000
AI 立刻返回:
评级:B 级 CPM:71.4(行业基准 15,当前报价明显虚高) 建议报价:¥600 话术:「您好,我们参考同类账号数据,这个体量的合理区间在 ¥500-700, 您这边能接受吗?长期合作的话我们优先返单。」
不是给你一堆数据让你自己分析,是直接告诉你结论 + 下一步动作。
6 个 MCP 工具
定价标准(内置)
这套评级体系是整个系统的核心,解决了「不知道报价贵不贵」的问题:
有报价时,按 CPM 判断:
无报价时,按粉赞比评估达人质量:
快速接入
git clone https://github.com/sz8887031-bot/kol-claw-mcp.git cd kol-claw-mcp pip install -r requirements.txt # 注册到 Claude Code claude mcp add openclaw -- python /path/to/kol-claw-mcp/mcp_server.py
自定义数据目录:
OPENCLAW_DATA_DIR=/your/data/path claude mcp add openclaw -- python /path/to/mcp_server.py
3 个斜杠命令
在 Claude Code 中直接使用,无需额外配置:
数据格式
默认读取 data/达人跟进表.csv,data/sample/ 提供完整演示数据:
cp data/sample/*.csv data/
aae317271d85OBSERVED · 2026-10-08Exposed tools (6)
4 read · 2 write · 0 destructive.
| Tool | Risk | Description |
|---|---|---|
add_creator | write | |
analyze_creator | read | |
generate_outreach_script | read | |
get_daily_tasks | read | |
list_creators | read | |
update_creator_status | write |
Trust audit
SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | PASS |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | PASS |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- none-observed
- Network
- none-observed
- Shell
- none-observed
- Dependencies
- not all pinned
- Secrets in source
- none-found
Findings (1)
mcp
Gates applied: no_behavioural_pass.
aae317271d85full audit observations/trust-audit/mcp-server/sz8887031-bot__kol-claw.json · Report an issue / request a re-scanAudit history
Every audit this server has had. A grade with a past is a grade somebody is still checking.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-08 | aae317271d85 | SAFE | B | 89 | first audit |
Questions
What is the Kol Claw MCP server?
把 KOL 投放判断力接入任意 AI:达人评级、CPM 定价、话术生成,6 个 MCP 工具,直接给结论
What tools does Kol Claw expose?
6 in total: 4 read-only, 2 that write, and 0 that can delete or overwrite. Every one is listed on this page with its risk.
Is Kol Claw safe to connect to an agent?
The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean server reads B.
What credentials does Kol Claw need?
No credential environment variables were found in its source, so it appears to need none.
How current is this page?
The grade is for one exact copy of the source (aae317271d85), read on 2026-10-08. The repository is watched and re-audited when it changes.