← Back to Skills Marketplace
174
Downloads
0
Stars
0
Active Installs
8
Versions
Install in OpenClaw
/install veevid
Description
AI video generator — text to video, image to video, and reference-to-video via Veevid API. Supports Kling 3.0, Sora 2, Veo 3.1, LTX 2.3, Seedance, Wan 2.6, a...
Usage Guidance
This skill appears to do exactly what it says: it uses a Veevid API key you store at ~/.config/veevid/api_key and calls veevid.ai endpoints to quote and generate videos. Before installing, consider: (1) the agent will read recent channel messages to locate images for image-to-video requests and will post quoted previews back to the channel — ensure you're comfortable with that privacy behavior in group chats; (2) keep your Veevid API key private (the skill reads it from a local file and uses it in requests); (3) there is no installer, so only the config file is created; and (4) the registry metadata contains a minor formatting error for the config path (harmless but worth noting). If you plan to use this skill in sensitive channels, limit where it can be invoked or remove agent permissions to read channel history.
Capability Analysis
Type: OpenClaw Skill
Name: veevid
Version: 1.2.2
The Veevid skill is a legitimate integration for an AI video generation service. It provides comprehensive instructions for the agent to handle text-to-video and image-to-video requests using the Veevid API (veevid.ai). Notable features include a cost-estimation step (quoting) and a user confirmation requirement before credit-consuming actions are taken. The image retrieval logic for chat environments is well-defined and includes privacy considerations, such as filtering attachments by the sender's ID. No evidence of data exfiltration, malicious execution, or harmful prompt injection was found.
Capability Assessment
Purpose & Capability
The name/description (Veevid text/image/reference-to-video) align with the declared behavior: it expects a Veevid API key stored at ~/.config/veevid/api_key and calls veevid.ai endpoints. The only minor metadata mismatch: registry metadata shows 'Required config paths: [object Object]' which looks like a formatting error, but the SKILL.md clearly documents the API key path.
Instruction Scope
Runtime instructions are narrowly scoped to: (1) read a local API key file, (2) call veevid.ai endpoints (quote/generate/status/credits/storage), and (3) optionally read recent chat messages to find image attachments when the user requested image-to-video. Reading recent messages and posting quoted previews is justified for the I2V flow, but it is a privacy-sensitive action — the agent will access channel history and post previews back to the channel. The skill does not instruct reading other local files or sending data to third-party endpoints beyond veevid.ai.
Install Mechanism
Instruction-only skill with no install spec and no code to write to disk beyond the expected config directory (~/.config/veevid). This is low-risk from an installation perspective.
Credentials
The skill requests a single local config path (~/.config/veevid/api_key) for the Veevid API key and uses that key in Authorization headers. No unrelated environment variables, credentials, or broad system access are requested.
Persistence & Privilege
No 'always' privilege is requested; the skill runs only when invoked. It stores/reads its own config file under the user's home config directory, which is expected for an external API integration. It doesn't modify other skills or global agent settings.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install veevid - After installation, invoke the skill by name or use
/veevid - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.2.2
Remove internal _meta.json and .clawhub from published package.
v1.2.1
Revert Chinese keywords from description/keywords (keep English only in metadata). Content matches v1.1.0.
v1.2.0
Add Chinese trigger phrases (图片生视频/文字生视频/生成视频 etc.) for better skill matching in Chinese conversations.
v1.1.0
Install via ClawHub only (GitHub repo now private). Fix Grok native audio, remove Kling 3.0 4K claim, add credits unit.
v1.0.3
Add keywords array to frontmatter for improved search discoverability (text to video, image to video, AI video generator, etc.)
v1.0.2
SEO: optimize title and description for ClawHub search — add text-to-video, image-to-video, AI video generator keywords
v1.0.1
Declare config paths (api_key) in SKILL.md frontmatter to resolve ClawHub security scanner false positive
v1.0.0
Initial release: 10 AI video models, text/image/reference-to-video, quote-before-generate flow, complete API reference
Metadata
Frequently Asked Questions
What is Veevid AI Video Generator?
AI video generator — text to video, image to video, and reference-to-video via Veevid API. Supports Kling 3.0, Sora 2, Veo 3.1, LTX 2.3, Seedance, Wan 2.6, a... It is an AI Agent Skill for Claude Code / OpenClaw, with 174 downloads so far.
How do I install Veevid AI Video Generator?
Run "/install veevid" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Veevid AI Video Generator free?
Yes, Veevid AI Video Generator is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Veevid AI Video Generator support?
Veevid AI Video Generator is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Veevid AI Video Generator?
It is built and maintained by Veevid AI (@meigesir); the current version is v1.2.2.
More Skills