# Gravity Skill

The **Gravity Skill** is an official [Agent Skill](https://docs.claude.com/en/docs/claude-code/skills) that loads Gravity's chain knowledge into your AI coding assistant on demand — network params, the system-contract address map, the native on-chain oracle, the cross-chain G token bridge, safe on-chain randomness, and the canonical EVM preinstalls.

It's written in the standard [Agent Skills](https://github.com/anthropics/agent-skills) (`SKILL.md`) format, so any agent that understands skills — **Claude Code, Cursor, Codex, and 50+ others** — can use it. Source: [Galxe/gravity-skills](https://github.com/Galxe/gravity-skills).

## Install

The easiest way is the [`skills`](https://github.com/vercel-labs/skills) CLI, which installs into whichever agent you use (Claude Code, Cursor, Codex, OpenCode, Cline, Copilot, and more):

```bash
npx skills add https://github.com/Galxe/gravity-skills
```

Then ask your agent anything about building on Gravity and it loads the skill on demand. Re-run `npx skills update` to pull the latest.

{% hint style="info" %}
**Claude Code plugin** — add the marketplace and install:

```
/plugin marketplace add Galxe/gravity-skills
/plugin install gravity@gravity-skills
```

**Manual** — point any `SKILL.md`-aware agent at the [`skills/gravity/`](https://github.com/Galxe/gravity-skills/tree/main/skills/gravity) directory, or copy it into your agent's skills folder.
{% endhint %}

## What it covers

| Skill     | What it covers                                                                                                                                                                                                                                                                                                                                        |
| --------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| `gravity` | Core reference for writing, deploying, and integrating smart contracts on Gravity L1: network parameters, the system-contract address map, the native oracle ABI, the cross-chain G token bridge (including a copy-paste `cast` permit recipe), safe on-chain randomness, and the canonical EVM preinstalls (Multicall3, Permit2, CreateX, ERC-4337). |

Each skill uses progressive disclosure: a short `SKILL.md` entry point that links into focused `references/` docs and runnable `examples/` (e.g. `OracleConsumer.sol`, `RandomnessConsumer.sol`, and the G-bridge walkthrough).

## Sources of truth

Addresses and ABIs in the skill are distilled from the canonical Gravity repositories. When in doubt, defer to:

* Core system contracts: [Galxe/gravity\_chain\_core\_contracts](https://github.com/Galxe/gravity_chain_core_contracts)
* This documentation site: [docs.gravity.xyz](https://docs.gravity.xyz)

{% hint style="warning" %}
Addresses are load-bearing. The skill instructs agents to copy hex addresses verbatim and to surface `TBA` / `blocked` statuses rather than inventing values.
{% endhint %}

## See also

* [Deployed Contracts (EVM Preinstalls)](/developer-resources/deployed-contracts.md) — the canonical addresses the skill references.
* [System Contracts](/developer-resources/mainnet.md) — Gravity's system runtime and bridge contracts.
* [Gravity Mainnet (L1)](/gravity-networks/l1-mainnet.md) — network connection parameters.


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.gravity.xyz/developer-resources/gravity-skill.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.