← Back to Skills Marketplace
Video Searching
by
memories-ai-official
· GitHub ↗
· v1.0.2
· MIT-0
501
Downloads
3
Stars
1
Active Installs
3
Versions
Install in OpenClaw
/install video-searching
Description
Search and analyze videos across YouTube, TikTok, Instagram, and X/Twitter via the Memories.ai Video Searching API.
Usage Guidance
This skill appears to implement what it claims, but check a few things before installing: 1) confirm the runner path the agent will execute — SKILL.md references scripts/run_video_query.sh but the bundle contains run_video_query.sh at the root (fix or reconcile the path). 2) Verify the API host (mavi-backend.memories.ai) is the official backend for the vendor linked on the homepage, and confirm the expected Authorization header format (the script sends the raw token). 3) Ensure MEMORIES_API_KEY you provide has the minimal scope needed and that logs/agent output won't leak the token. 4) If you don't trust the skill source, require manual review or run the script in an isolated environment first.
Capability Analysis
Type: OpenClaw Skill
Name: video-searching
Version: 1.0.2
The video-searching skill is a legitimate tool designed to interface with the Memories.ai Video Searching API. The implementation in run_video_query.sh safely handles user input using jq's argument passing to prevent shell or JSON injection. It correctly manages API authentication via the MEMORIES_API_KEY environment variable and processes Server-Sent Events (SSE) into structured NDJSON for the agent, with no evidence of malicious behavior or data exfiltration.
Capability Tags
Capability Assessment
Purpose & Capability
Name/description, required binaries (curl, jq), and required env var (MEMORIES_API_KEY) line up with a video-search API integration. However, the SKILL.md calls a runner at <skill_dir>/scripts/run_video_query.sh while the included file is run_video_query.sh at the repository root (path mismatch). The API host used by the script (mavi-backend.memories.ai) differs from the public homepage domain (api-tools.memories.ai) — that can be normal but is worth checking.
Instruction Scope
SKILL.md runtime instructions stay within the task: construct query JSON, call the SSE endpoint, parse NDJSON events, and present results / clarifications / errors. The instructions do not ask the agent to read unrelated files or additional environment variables. The only scope issue is the referenced script path that doesn't match the packaged file name.
Install Mechanism
This is an instruction-only skill with one included shell runner; there is no download/install step or external package fetch. That minimizes install-time risk.
Credentials
Only MEMORIES_API_KEY is required, which is appropriate for this API-based skill. The script sends the key in an Authorization header; verify the service expects the token in that exact header format and avoid logging the value.
Persistence & Privilege
Skill is user-invocable, not always: true, and can be invoked autonomously per platform default. It does not request system-wide config or other skills' credentials.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install video-searching - After installation, invoke the skill by name or use
/video-searching - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.2
No functional changes in this version.
- No file changes were detected for version 1.0.2.
- Behavior and documentation remain unchanged.
v1.0.1
## v1.1.0 — Migrated to Memories.ai Managed API
### Breaking Changes
- Replaced self-hosted Python agent with managed Video Searching API (`api-tools.memories.ai`)
- Changed invocation command from `/video_sourcing` to `/video_search`
- New authentication: `MEMORIES_API_KEY` replaces `GOOGLE_API_KEY` + `YOUTUBE_API_KEY`
### Improvements
- Zero cold-start — no repo cloning or dependency installation needed
- Simplified requirements: only `curl` and `jq` (removed `git`, `uv`, Python)
- SSE streaming with real-time progress updates
- Enhanced output format: platform badges (🎬🎵📸🐦), relevance notes, and key video metrics
- Added support for `--platforms` and `--time-frame` parameters
- Improved error handling and clarification messaging
v1.0.0
Initial release: deterministic video searching agent with self-bootstrap runtime
Metadata
Frequently Asked Questions
What is Video Searching?
Search and analyze videos across YouTube, TikTok, Instagram, and X/Twitter via the Memories.ai Video Searching API. It is an AI Agent Skill for Claude Code / OpenClaw, with 501 downloads so far.
How do I install Video Searching?
Run "/install video-searching" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Video Searching free?
Yes, Video Searching is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Video Searching support?
Video Searching is cross-platform and runs anywhere OpenClaw / Claude Code is available (darwin, linux).
Who created Video Searching?
It is built and maintained by memories-ai-official (@memories-ai-official); the current version is v1.0.2.
More Skills