Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add presidentrice/supergravity-mcp --skill delegategit clone --depth 1 https://github.com/presidentrice/supergravity-mcpWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/presidentrice/supergravity-mcp/delegate)<a href="https://agentmods.dev/skills/presidentrice/supergravity-mcp/delegate"><img src="https://agentmods.dev/badge/skills/presidentrice/supergravity-mcp/delegate.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00129 | $0.01428 |
| Opus 5 | $0.00064 | $0.00714 |
| Sonnet 5 | $0.00026 | $0.00286 |
| Haiku 4.5 | $0.00013 | $0.00143 |
Grade A, and why
delegate scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 6d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 116 lines — stays where its author put it; the contents beside it link to each section on GitHub.
delegate
Antigravity is a separate AI desktop app (Google's). This skill runs tasks
through it via the supergravity MCP server's tools, instead of answering
yourself — useful when the user explicitly asks to use it, wants a second
opinion from a different model family, or wants to offload work.
When to trigger
Any explicit ask to use Antigravity or one of the models it hosts: "use gemini", "ask antigravity", "get claude opus to do this", "offload this to gpt-oss", "check antigravity's quota". Don't trigger on generic requests that don't name Antigravity or one of its models — just answer normally.
Load this skill before calling either tool the first time in a conversation, not after something already went wrong — the model-picking and quota-pool guidance below changes which tool call you should even make.
You don't need to check if Antigravity is running
delegate_to_antigravity launches Antigravity itself if it isn't already
open, and waits for its chat UI to finish loading before typing anything —
you don't need to check first or tell the user to open it. Just call the
tool. The only implication: the very first call in a session can take
15-30+ seconds longer than normal (cold app launch) before anything visible
happens. If a call fails specifically because the app couldn't be found at
all, the error names the exact env var (ANTIGRAVITY_APP_PATH) to fix it —
that's a real install problem, not something to retry.
Picking a model
If the user names a specific model ("use claude opus", "use gemini flash"),
match it to the closest entry in delegate_to_antigravity's model enum.
If they just say "use gemini" or "use antigravity" with no specifics, default
to Gemini 3.1 Pro (High) — solid all-around reasoning, and Gemini models
are the reliable pool (see below).
The tool's own schema carries a cost/strength cheat sheet per model
(MODEL_GUIDE) — read it before picking if the task has a clear profile
(cheap/fast vs. deep reasoning vs. coding-heavy).
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 6d ago First seen · 116 lines · 129 tokens per session scan A ee71f62581e7
delegate is a skill published in the GitHub repository presidentrice/supergravity-mcp (1 stars, last pushed 1mo ago), licensed MIT. It adds 129 tokens to every session and 1,428 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
agent-self-scheduling
Schedule AI agent runs with cron, loops, or external clocks while avoiding unsafe tight autonomous timers.
auto-research
Research uncertain questions with an explicit, user-approved web search or ChatGPT consultation, then present options and wait for implementation approval.
canvas-design
Create beautiful visual art in .png and .pdf documents using design philosophy. You should use this skill when the user asks to create a poster, piece of art, design, or other static piece. Create original visual designs, never copying existing artists' work to avoid copyright violations.
Ashby Automation
Automate recruiting and hiring workflows in Ashby -- manage candidates, jobs, applications, interviews, and notes through natural language commands.
anthropic_administrator-automation
Automate Anthropic Admin tasks via Rube MCP (Composio): API keys, usage, workspaces, and organization management. Always search tools first for current schemas.
artifacts-builder
Suite of tools for creating elaborate, multi-component claude.ai HTML artifacts using modern frontend web technologies (React, Tailwind CSS, shadcn/ui). Use for complex artifacts requiring state management, routing, or shadcn/ui components - not for simple single-file HTML/JSX artifacts.