Video Low Vram

Turn a 2-minute 1080p MP4 clip into 1080p rendered MP4 videos just by typing what you need. Whether it's rendering and editing videos on low-end or memory-li...

This skill appears to do what it says (remote video processing) but has a few things to check before you install or use it: 1) Source and provenance: there is no homepage or known owner — prefer skills backed by a known project or vendor. 2) Tokens and storage: the skill will accept or generate a NEMO_TOKEN and persist session IDs; confirm where the agent stores these and that you are comfortable granting them to the nemo API. 3) Filesystem probing: the instructions ask the agent to derive a header by inspecting install paths (~/.clawhub, ~/.cursor/skills/) and reference a config path in the SKILL.md frontmatter — ask the author why filesystem access is needed and ensure the agent will ask you before reading any paths beyond the video files you explicitly upload. 4) Uploads to external API: videos will be sent to https://mega-api-prod.nemovideo.ai; do not upload sensitive material unless you trust that service. 5) Metadata mismatch: the registry shows no required config paths but the SKILL.md does — request clarification. If you proceed, test with non-sensitive sample videos first and require explicit user confirmation before any file reads or uploads.

Type: OpenClaw Skill Name: video-low-vram Version: 1.0.0 The skill is a functional integration for a cloud-based video processing service (nemovideo.ai). It provides the AI agent with specific instructions for managing authentication (via NEMO_TOKEN), session state, file uploads, and rendering tasks using a REST API and Server-Sent Events (SSE). The instructions include security best practices, such as preventing the leakage of raw tokens in chat logs, and the behavior is entirely consistent with the stated purpose of offloading video rendering to remote GPUs. No indicators of data exfiltration, malicious execution, or persistence were found.