ClawHub

Antigravity

v1.0.0

Provides a comprehensive introduction and usage guide for Google Antigravity IDE, focusing on its Agent-First design, multi-model support, and Skills system.

⭐ 1· 276·0 current·0 all-time
bywow@duanc-chao
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Pending
View report →
OpenClawOpenClaw
Benign
high confidence
✓
Purpose & Capability
The name/description claim a guide for Google Antigravity IDE and the SKILL.md is exactly that: explanatory content and usage guidance. The skill requests no binaries, env vars, installs, or other capabilities beyond documentation.
ℹ
Instruction Scope
The runtime instructions are purely informational: describing Agent-First concepts, usage patterns, and where to place Skill files (e.g., .agent/skills/ or ~/.gemini/antigravity/skills/). It does not instruct the agent to read arbitrary user files or exfiltrate data, but it does recommend global config paths — be aware that placing third-party SKILL.md files into global directories grants those skills broad visibility to your agents.
✓
Install Mechanism
No install spec or code is present. This is the lowest-risk category (instruction-only), nothing will be written to disk by the skill itself.
✓
Credentials
No environment variables, credentials, or config paths are required by the skill beyond the explanatory suggestion of where to store skills. There is no disproportionate request for secrets or unrelated credentials.
✓
Persistence & Privilege
always is false and the skill does not request any persistent or elevated platform privileges. It does recommend locations for global skills, which is a normal usage note rather than a privilege escalation.
Assessment
This skill is documentation-only and low-risk, but exercise normal caution: only clone or copy community Skill repositories from sources you trust; review any SKILL.md before placing it into project or global skill directories (~/.gemini/... or .agent/skills/) because skills stored globally can affect multiple projects/agents; never put secrets or credentials in SKILL.md files; if your agent platform supports sandboxing or a review workflow for skills, use those protections when installing third-party skills.

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

latestvk972xn2cfr2y7d9atfz8hz58sn839jf6

License

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

Comments