ClawHub

Which Sparki Skill Should I Use?

v1.0.0

AI-powered video editing — turn raw footage into polished vlogs, highlight reels, TikTok/Shorts/Reels, montages, and more. Control the editing flow through T...

⭐ 0· 94·0 current·0 all-time
by@fischerlam
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
high confidence
✓
Purpose & Capability
Name/description are a routing/helper for Sparki editing. Declared primary credential (SPARKI_API_KEY), network host (agent-api.sparki.io), filesystem writes for config and output, and CLI commands to upload/edit/download all match the stated purpose.
✓
Instruction Scope
SKILL.md instructs only to validate an API key, provide upload links, accept local file paths or Telegram Mini App uploads, and run the included sparki CLI commands. It does not instruct reading unrelated system files or exfiltrating data to unexpected endpoints.
✓
Install Mechanism
This is effectively instruction-only with bundled Python CLI sources; no remote downloads from arbitrary URLs. The metadata references a 'uv sync' step requiring a 'uv' binary — unusual but not inherently risky; it is local and not a network fetch to an unknown host.
ℹ
Credentials
Primary credential SPARKI_API_KEY is appropriate. Minor mismatch: code reads SPARKI_UPLOAD_TG_LINK as an optional override but that env var is not listed in metadata env array; metadata otherwise lists no unrelated secrets. Base URL can be overridden in config/CLI which could point to other hosts, but the skill's declared network permission is limited to agent-api.sparki.io.
✓
Persistence & Privilege
always:false (no forced inclusion). The skill writes only its own config (~/.openclaw/config) and outputs (~/.openclaw/workspace/sparki/videos), which is appropriate for a CLI that saves API keys and downloads results. It doesn't modify other skills or system-wide settings.
Assessment
This package appears to be a legitimate CLI wrapper for Sparki's editing service. Before installing, consider: (1) it requires/store a Sparki API key — you can provide it via SPARKI_API_KEY or the CLI 'sparki setup', and the key will be saved under ~/.openclaw/config/sparki.json; only provide a key you trust Sparki with. (2) It will upload video files to agent-api.sparki.io and download results to ~/.openclaw/workspace/sparki/videos — ensure you are comfortable with that network destination and file locations. (3) The metadata requires a 'uv' binary and runs 'uv sync' during install; confirm what 'uv' is in your environment. (4) The code optionally reads SPARKI_UPLOAD_TG_LINK env var (not explicitly listed in metadata) and allows overriding base_url in config — if you plan to change the base_url, be aware network behavior may change. If you trust sparki.io and the official Telegram bot link referenced, the skill is coherent with its stated purpose.

Like a lobster shell, security has layers — review code before you run it.

discoveryvk974dq17kj2nzmv32rdp6awe0x83fggrguidevk974dq17kj2nzmv32rdp6awe0x83fggrlatestvk974dq17kj2nzmv32rdp6awe0x83fggrsparkivk974dq17kj2nzmv32rdp6awe0x83fggr

License

MIT-0
Free to use, modify, and redistribute. No attribution required.
Termshttps://spdx.org/licenses/MIT-0.html

Runtime requirements

🎬 Clawdis
OSmacOS · Linux
Binsuv
Primary envSPARKI_API_KEY

Comments