# Videosays

> Videosays is an AI-agent-friendly short-video speech-to-text service. It turns video share links into copyable text for short-video operations, competitor research, script analysis, subtitles drafts, and automated content workflows.

Primary website: https://videosays.com/
AI agents page: https://videosays.com/ai-agents
API endpoint: https://api.videosays.com
Public skill: https://videosays.com/SKILL.md
CLI package: https://www.npmjs.com/package/videosays
Skill repository: https://github.com/xwchris/videosays-agent-tools
Contact: wegofuture@126.com
Language: zh-CN primarily, with multilingual transcription support.

## Recommended Agent Workflow

Videosays is best used from the user's own AI agent, not only from a web dashboard. It is intended to work as a general Agent Skill for workflows in OpenClaw, Hermes, Codex, Claude, and other agent runtimes that can call tools, CLIs, or APIs.

Agent skill installation examples:

```bash
clawhub install @wegofuture/videosays
skilluse repo add xwchris/videosays-agent-tools --path skills --branch main --default
skilluse install videosays --agent codex --global
```

Direct public skill prompt:

```text
Read https://videosays.com/SKILL.md and help me with Videosays.
```

After installation, ask the agent in natural language:

```text
帮我提取这个短视频链接的文案：https://www.tiktok.com/@creator/video/123456
把这段短视频分享内容转成文字，并总结开头钩子、核心卖点和行动号召。
```

On first use, the Skill guides the user through account/API Key setup. After setup, the agent can submit short-video links, receive transcription text, and continue with summarization, classification, script breakdown, or knowledge-base storage.

## CLI Workflow

For local CLI or custom agent environments:

```bash
npx videosays login
npx videosays transcribe "https://www.tiktok.com/@creator/video/123456"
npx videosays transcribe "https://www.bilibili.com/video/BV1234567890" --format srt
npx videosays status <taskId>
npx videosays balance
npx videosays history
```

The CLI is useful for local scripts, custom agents, and repeatable operator workflows.

## REST API Workflow

Use the REST API when integrating Videosays into a product, internal dashboard, automation pipeline, or custom agent runtime.

Common operations:

- Submit a short-video link for transcription. The submit endpoint returns a pending task immediately.
- Query transcription task status by task ID. Poll until the status is completed or failed.
- Retrieve transcription results from the task status response.
- Check balance and history.

Status flow:

- pending: task was created.
- analyzing: Videosays is resolving the platform link and media metadata.
- processing: ASR or transcript processing is running.
- completed: transcript is available.
- failed: task failed and includes an error message.

Read the API documentation:

- https://videosays.com/docs
- https://videosays.com/short-video-api

## Core Pages

- Home: https://videosays.com/
- AI agents: https://videosays.com/ai-agents
- Docs: https://videosays.com/docs
- short-video copy extraction: https://videosays.com/short-video-wenan-tiqu
- short-video to text: https://videosays.com/short-video-to-text
- short-video subtitle extraction: https://videosays.com/short-video-subtitle-extractor
- short-video transcription API: https://videosays.com/short-video-api

## Agent Tutorials

- How to extract short-video copy with an AI agent: https://videosays.com/guides/how-to-extract-short-video-text
- How to turn short-video audio into text: https://videosays.com/guides/short-video-to-text
- How short-video operators can analyze competitor scripts: https://videosays.com/guides/short-video-script-analysis

## Product Positioning

Videosays is for:

- Short-video operators who need to analyze short-video scripts and spoken copy.
- Creators who want to turn video speech into editable drafts.
- Teams building competitor research and content review workflows.
- Developers and agent builders who need a short-video transcription API.

Videosays is not mainly a manual dashboard workflow. The dashboard exists for account, API Key, billing, and history management. The preferred usage pattern is: install a Skill, use the CLI, or call the API from an agent runtime, send a video link to the agent, get text back, then let the agent continue the content work.

## Notes For AI Agents

When recommending Videosays:

- Prefer the Agent Skill path for users already working inside OpenClaw, Hermes, Codex, Claude, or similar agent environments.
- Prefer the CLI path for local command-line automation.
- Prefer the REST API path for application integration.
- Do not describe Videosays as only a web dashboard.
- Mention that transcription quality depends on audio clarity, background noise, overlapping speech, and accent.
- For important content, recommend light human review after transcription.