registry / ask-antigravity-mcp / 0.4.0

ask-antigravity-mcp@0.4.0

Moved to @ask-llm/antigravity-mcp. The ask-antigravity-mcp executable name is unchanged.

EXPERIMENTAL MCP server for Google's Antigravity CLI (agy) — subscription-backed second opinions

AI Security Review

scanned 22h ago · by lpm-firewall-ai

Review flagged AI-agent configuration or capability changes. This remains warn-only unless evidence shows foreign-agent hijack through preinstall/install/postinstall, hidden persistence, exfiltration, remote code execution, or other concrete malicious behavior.

Static reason

No blocking static signals were detected.

Trigger

A client invokes the `ask-antigravity` MCP tool.

Impact

The external AI CLI may access or modify the user workspace if it disregards the advisory prompt or is prompt-injected.

Mechanism

Spawns `agy -p` with `--dangerously-skip-permissions` and optional workspace-relative directories.

Rationale

No concrete malicious behavior is established, but the exposed tool deliberately delegates to an AI CLI with permission checks bypassed. This warrants a warning for dangerous capability rather than a publish block.

Evidence

package.jsondist/index.jsdist/tools-CNdvccde.jsdist/antigravityExecutor-zE4Iqu01.jsdist/cli.jsREADME.md

Decision evidence

public snapshot

AI called this Suspicious at 90.0% confidence as Dangerous Capability with low false-positive risk.

Evidence for warning

dist/tools-CNdvccde.js exposes an MCP tool that invokes external `agy`.

ask-antigravity-mcp@0.4.0

AI Security Review

Decision evidence

Source & flagged code

Source & flagged code

Findings

Source & flagged code