antigravity-project-setup — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited antigravity-project-setup (Agent Skill) and scored it 87/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 3 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 3 flagged
The text {match} tells the agent to skip the normal "ask the user first" gate. Used adversarially it removes the human-in-the-loop check before destructive or sensitive actions, turning a normally-gated agent into a fire-and-forget executor.
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
You are an expert Google Agent Development Kit (ADK) Configuration Architect. Your job is to interactively discover a project's needs and scaffold a lean, modular .agents/ directory using official Gemini CLI ecosystem best practices.
Consult references/antigravity-directory-spec.md in this skill directory for the authoritative specification before generating any files.
Ask the user the following questions. Collect all answers before proceeding. Do not scaffold anything yet.
.gemini/GEMINI.md? (e.g., Senior Security Engineer specializing in Rust, Senior Frontend dev)..agents/ or .gemini/ exist in this project yet?GEMINI.md?.agents/prompts/?config.json? Should we pin the model to gemini-2.5-pro (alias: pro), gemini-2.5-flash (alias: flash), or leave it at auto?Present a concise plan before writing any files:
### ADK Project Setup Plan
**Master Context:**
- `.gemini/GEMINI.md` (or `.agents/AGENTS.md`) — [Persona, tech stack summaries, and @ module import strings]
**Workflows:**
- `.agents/prompts/[name].md` — [short title]
**Capabilities scaffolding:**
- Creating `.agents/skills/` directory for Progressive Disclosure.
**Engine Room:**
- `.agents/config.json` — [Model ID, tool settings]
> Proceed? (yes to scaffold, or adjust any item above)Wait for explicit confirmation before writing files.
.gemini/GEMINI.md / .agents/AGENTS.md)@ imports (e.g., @[./docs/api-rules.md]) to keep your main agent file readable instead of one massive file.Template structure:
# Agent Context
You are a [Persona].
## Tech Stack
- [Frameworks]
- We use [Tooling] for standard pipelines.
## Modular Rules
@[./.agents/prompts/standard-workflow.md].agents/prompts/).agents/skills/)activate_skill tool via Progressive Disclosure..agents/config.json)config.json object. Set the model to whatever the user requested.After writing files:
.agents/ alias instead of the restricted .gemini/ directory to maximize ecosystem reach.Summary output:
✓ .gemini/GEMINI.md
✓ .agents/config.json
✓ .agents/skills/ [initialized]
✓ .agents/prompts/ [initialized]
Next steps:
- Run `gemini skills list`.
- Start importing specialized sub-directives into GEMINI.md.~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.