Video Searching
PassAudited by ClawScan on May 1, 2026.
Overview
The skill is coherent for video search, but users should notice that it sends search queries to Memories.ai using an API key and has a minor runner-path mismatch in the instructions.
This looks safe for its stated purpose if you trust Memories.ai with your video-search queries and API-key usage. Before installing, confirm the runner path is correct, use a dedicated API key if possible, and avoid putting confidential information into search prompts.
Findings (4)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
Searches may be performed under the user's Memories.ai account and may consume quota or be logged by the provider.
The runner authenticates API calls with the user's MEMORIES_API_KEY. This is expected for the Memories.ai integration, but it is still delegated account authority.
-H "Authorization: ${MEMORIES_API_KEY}"Use a dedicated or scoped API key if available, keep it out of shared environments, and monitor provider usage.
Any private or sensitive terms included in a video search query may be shared with Memories.ai.
The artifacts clearly disclose that user queries are sent to an external API provider and results are streamed back.
This skill calls the Memories.ai Video Searching API — a managed, token-authenticated endpoint that searches across platforms and returns structured results via SSE.
Avoid including confidential information in search queries unless the provider's privacy and retention terms are acceptable.
The agent can launch the included query runner during matching video-search tasks.
The skill directs the agent to execute a local runner process and poll it. This is expected for streaming API results and is not shown to perform unrelated commands.
Start with `exec` using `background: true`.
Install only if you are comfortable with the agent invoking this local runner for video search requests.
The skill may fail to run until the runner path is corrected or confirmed.
The instruction references a scripts/ path, while the provided artifact manifest lists run_video_query.sh at the root. This appears to be a packaging or documentation mismatch rather than hidden behavior.
<skill_dir>/scripts/run_video_query.sh
Verify the installed runner path before use and prefer a package version whose instructions and manifest agree.