antigravity-maintainer-batch-release

Pass

Audited by Gen Agent Trust Hub on Jul 25, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill requires the execution of multiple repository-specific scripts defined in package.json (e.g., npm run merge:batch, npm run release:publish) and uses the aas CLI tool to modify local MCP host configurations.
  • [EXTERNAL_DOWNLOADS]: The skill manages the agentic-awesome-skills package from npm and references GitHub Pages sites (sickn33.github.io). These are vendor-owned resources and are standard for the tool's intended purpose.
  • [PROMPT_INJECTION]: The skill instructs the agent to read and evaluate live PRs, issues, and discussions on GitHub. This ingestion of untrusted external content creates a surface for indirect prompt injection.
  • Ingestion points: Reads PR descriptions, comments, and issue discussions from GitHub (SKILL.md).
  • Boundary markers: Absent; there are no specific instructions to ignore or treat embedded commands as data.
  • Capability inventory: Includes shell command execution via npm, local file system modifications, and management of MCP configurations (SKILL.md).
  • Sanitization: Relies on manual maintainer review rather than automated filtering or escaping of the ingested external content.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 25, 2026, 02:58 PM
Security Audit — agent-trust-hub — antigravity-maintainer-batch-release