← 返回 Skills 市场
Openclaw Talk Analyzer
作者
Justin Liu
· GitHub ↗
· v1.0.0
· MIT-0
333
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install talk-analyzer
功能描述
基于 AI 的对话分析工具，自动提取会议要点、销售异议、客户满意度、行动项及策略建议，支持多场景应用。
安全使用建议
This skill appears to do what it says (conversation analysis), but the package metadata omits important operational details found in the SKILL.md/README. Before installing or using it: 1) Verify the source repository (the README links to a GitHub repo) and inspect the code there — do not blindly run npm install from an unknown repo. 2) Expect to provide an AI API key (Anthropic/OpenAI) or run a local LLM; treat those keys as sensitive and do not expose them to untrusted code or services. 3) The skill assumes a CLI/binary (openclaw-talk) that is not bundled — confirm how that binary is delivered and who maintains it. 4) If your transcripts are sensitive, prefer the documented local LLM option and audit any network calls the tool makes. 5) If you rely on an organization security policy, get approval before installing external packages or supplying API keys.
功能分析
Type: OpenClaw Skill
Name: talk-analyzer
Version: 1.0.0
The skill bundle contains documentation and metadata for a conversation analysis tool. The instructions in skill.md and readme.md are consistent with the stated purpose of using AI to summarize meetings and sales calls, and there are no signs of malicious code, data exfiltration, or prompt injection attacks.
能力评估
Purpose & Capability
The described purpose — analyzing conversations and producing summaries/action items — is coherent with the skill's instructions. However, the SKILL.md and README state explicit technical requirements (Node.js 18+, npm/pnpm, and at least one AI API key such as ANTHROPIC_API_KEY or OPENAI_API_KEY) while the registry metadata declares no required env vars or required binaries. That mismatch means the manifest does not accurately reflect what the skill actually needs to operate.
Instruction Scope
The instructions expect the presence of a CLI/programming package (openclaw-talk / openclaw-talk-analyzer) and tell the agent to read input transcript files and call external AI APIs (Claude/OpenAI) or a local LLM. Those behaviors are consistent with the stated purpose, but the skill also includes examples of cloning a GitHub repo and running npm install — yet there is no install spec and no bundled code. The agent instructions therefore assume installing or using external code/binaries that are not provided by the skill package itself, which is an incoherence and a deployment risk if followed blindly.
Install Mechanism
This is an instruction-only skill with no install spec or code files (lowest immediate risk from the registry). The README and SKILL.md refer to installing from GitHub or npm, but that would require fetching external code — the skill does not include or declare those steps. Absence of an install spec is not dangerous by itself, but combined with the instructions it means a user/agent would need to fetch and run third-party code; verify provenance before doing so.
Credentials
The skill's text explicitly requires at least one AI service API key (Anthropic/OpenAI) and suggests storing keys in a .env file (ANTHROPIC_API_KEY / OPENAI_API_KEY). Yet the registry metadata lists no required environment variables or primary credential. This is a substantive inconsistency: the runtime behavior requires secrets (API keys) but the manifest does not declare them, which could cause surprise and potential accidental exfiltration if the agent supplies keys without explicit declaration and user consent.
Persistence & Privilege
The skill does not request persistent presence (always:false), does not declare modifying other skills or system-wide config, and requests no special config paths. It simply describes running analysis on provided transcript files. No elevated persistence or privilege escalation is evident from the manifest.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install talk-analyzer - 安装完成后，直接呼叫该 Skill 的名称或使用
/talk-analyzer触发 - 根据 Skill 的参数说明提供必要输入，即可获得结构化输出
版本历史
v1.0.0
Talk Analyzer v1.0.0 初始发布
- 支持自动分析会议、销售、客服、谈判、面试及通用对话，提取要点、行动项及决策
- 提供摘要、情感分析、发言者画像等多种智能提取功能
- 支持销售专用分析：异议识别、购买信号和跟进建议
- 可选择多款 AI 引擎（Claude、OpenAI、或本地 LLM）进行分析，兼顾隐私和灵活性
- 命令行和编程接口双模式使用，支持批量处理和多输出格式（json/markdown）
- 附带详细用法、环境说明和典型应用场景，开源免费
元数据
常见问题
Openclaw Talk Analyzer 是什么？
基于 AI 的对话分析工具，自动提取会议要点、销售异议、客户满意度、行动项及策略建议，支持多场景应用。 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 333 次。
如何安装 Openclaw Talk Analyzer？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install talk-analyzer」即可一键安装，无需额外配置。
Openclaw Talk Analyzer 是免费的吗？
是的，Openclaw Talk Analyzer 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Openclaw Talk Analyzer 支持哪些平台？
Openclaw Talk Analyzer 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。
谁开发了 Openclaw Talk Analyzer？
由 Justin Liu（@zhenstaff）开发并维护，当前版本 v1.0.0。
推荐 Skills