← 返回 Skills 市场
Tellers
作者
Robin Guignard-Perret
· GitHub ↗
· v1.0.2
· MIT-0
143
总下载
1
收藏
0
当前安装
3
版本数
在 OpenClaw 中安装
/install tellers
功能描述
Create, edit, and share AI-generated videos using tellers.ai — an AI video platform that aggregates leading generation models (Kling, Veo, LTX, ElevenLabs, a...
安全使用建议
Before installing or using this skill, confirm the source and authenticity of the 'tellers' CLI and the 'tellers-ai' Homebrew tap (prefer official docs or GitHub releases). Expect to need a TELLERS_API_KEY even though the registry metadata omits it — verify where that key is stored and consider setting it only for the session (not in a shared ~/.zshrc) or using a dedicated, limited-permission key. Run the CLI in a sandbox or a throwaway environment first to observe behavior, and avoid exposing unrelated files. If you must add the API key to a profile, restrict file permissions and rotate the key if you stop using the skill. Finally, consider contacting the skill owner or choosing a skill with verifiable homepage/source before granting long-term access.
功能分析
Type: OpenClaw Skill
Name: tellers
Version: 1.0.2
The 'tellers' skill bundle provides instructions for an AI agent to interact with the Tellers.ai video generation platform via its CLI. The SKILL.md file outlines standard workflows for media uploads, video generation, and project exporting using the 'tellers' command-line tool. It utilizes OpenClaw's 'sessions_spawn' feature to handle long-running tasks in subagents, which is a legitimate architectural pattern for this environment. No evidence of data exfiltration, malicious execution, or harmful prompt injection was found.
能力评估
Purpose & Capability
The SKILL.md clearly supports uploading media, running long-running generation, and building result URLs using the tellers CLI — that aligns with the skill name and description. However, the runtime instructions require a TELLERS_API_KEY and to install a CLI via 'brew tap tellers-ai/tellers', while the registry metadata lists no required env vars or install steps. The API key is reasonable for the claimed purpose, but the metadata/instructions mismatch is an incoherence.
Instruction Scope
Instructions explicitly tell the agent to run local CLI commands that read local file paths (e.g., /path/to/footage), spawn isolated subagents for long-running uploads/generations, parse JSON outputs, and construct app URLs. Those actions are expected for a CLI-backed video service. The only noteworthy scope issue is the advice to persist an API key in shell profiles (~/.zshrc, ~/.bashrc), which broadens where a secret is stored and could increase exposure.
Install Mechanism
The skill is instruction-only (no install spec in registry) but tells users to run 'brew tap tellers-ai/tellers && brew install tellers'. That uses a third‑party Homebrew tap rather than a standard, verified package source. Installing from an unfamiliar tap can pull arbitrary binaries; this is higher risk than using an official brew formula or a well-known release host.
Credentials
SKILL.md requires TELLERS_API_KEY (export TELLERS_API_KEY=sk_...), which is proportionate to the CLI's API usage. However, the registry metadata declares no required env vars (discrepancy). The instruction to add the API key export to persistent shell profiles increases surface for accidental disclosure. No other unrelated secrets are requested.
Persistence & Privilege
The skill does not request platform-level persistence (always: false) and does not modify other skills. It does instruct the user to persist an API key in shell startup files, which is a user-side persistence decision and increases long-term exposure of the secret but is not a skill-owned privilege escalation.
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 tellers - 安装完成后，直接呼叫该 Skill 的名称或使用
/tellers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
v1.0.2
Update skill to use new brew package instead of having to compile the cli from source
v1.0.1
Updated skill
v1.0.0
Initial release — upload media, generate AI videos, share previews via tellers.ai
元数据
常见问题
Tellers 是什么？
Create, edit, and share AI-generated videos using tellers.ai — an AI video platform that aggregates leading generation models (Kling, Veo, LTX, ElevenLabs, a... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 143 次。
如何安装 Tellers？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install tellers」即可一键安装，无需额外配置。
Tellers 是免费的吗？
是的，Tellers 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Tellers 支持哪些平台？
Tellers 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 Tellers？
由 Robin Guignard-Perret（@yxdunc）开发并维护，当前版本 v1.0.2。
推荐 Skills