Latest release: v1.0.1Download zip
Capabilities
Compatibility
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
The registered `soul_compute` tool matches the stated purpose: it sends a message and personality settings to the disclosed Neshama API and returns personality/emotion guidance. Users should notice that this is an external prompt-generation service.
Instruction Scope
The plugin returns `context_for_llm`, suggested tone, temperature, and behavior tendency, which are intended to influence the agent's responses. This is disclosed and purpose-aligned, with no evidence of hidden override instructions.
Install Mechanism
There is no install script or shell execution, but the packaged runtime appears incomplete: `dist/index.js` imports `./soul-compute` while the manifest does not include `dist/soul-compute.js`, and `.clawhubignore` excludes it. This is more of a packaging/provenance issue than evidence of malicious behavior.
Credentials
The skill requests no local files, OS privileges, or user credentials, but it performs a disclosed network call to `https://api.neshama.pw/v1/soul/compute`.
Persistence & Privilege
The plugin activates on startup to register its tool, but the artifacts do not show background workers, autonomous persistence, local state storage, privilege escalation, or credential harvesting.
Scan Findings in Context
[suspicious.exposed_secret_literal] expected: The flagged literal is `nsh_public_beta_2026`, which the README/SKILL.md explicitly labels as a public key and the code uses as the default API key for the disclosed Neshama endpoint. This is not evidence of a leaked private user secret in the supplied artifacts.
Assessment
This skill appears benign, but only use it when you are comfortable sending the selected message content to Neshama's API and receiving remote-generated personality guidance back into the agent context.src/soul-compute.ts:37
File appears to expose a hardcoded API secret or token.
Patterns worth reviewing
These patterns may indicate risky behavior. Check the VirusTotal and OpenClaw results above for context-aware analysis before installing.Verification
Tags
Neshama Soul - OpenClaw Plugin
给 AI Agent 注入灵魂,赋予独特人格和真实情绪。
安装
openclaw plugins install @neshama/openclaw-soul
或从 GitHub 安装:
git clone https://github.com/Neshama-AI/openclaw-soul-plugin.git
cd openclaw-soul-plugin
openclaw plugins install .
功能
soul_compute 工具
注册一个工具,允许 Agent 调用 Neshama Soul API 获取人格提示词。
参数:
message(必需): 用户消息ocean(可选): OCEAN 人格配置traits(可选): 人格特征session_id(可选): 会话 ID
返回:
context_for_llm: 人格提示词emotion_state: 情绪状态suggested_temperature: 建议温度suggested_tone: 建议语气
OCEAN 人格模型
| 维度 | 默认值 | 说明 |
|---|---|---|
| Openness | 0.7 | 开放性:好奇心、创造力 |
| Conscientiousness | 0.6 | 尽责性:条理性、自律性 |
| Extraversion | 0.8 | 外向性:社交能力、活力 |
| Agreeableness | 0.7 | 宜人性:合作性、同理心 |
| Neuroticism | 0.3 | 神经质:情绪稳定性 |
API 配置
- 端点:
POST https://api.neshama.pw/v1/soul/compute - 公共 Key:
nsh_public_beta_2026
相关资源
| 资源 | 链接 |
|---|---|
| 官网 | https://neshama.pw |
| SoulCraft | https://neshama.pw/soulcraft |
| API 文档 | https://api.neshama.pw/docs |
MIT License
© 2026 Neshama AI. All rights reserved.
