galleonlabs/crypto-defi-skills / packages/lending/skills/galleon-aave-v4
galleon-aave-v4
VerifiedUse when inspecting an Aave V4 Hub/Spoke position, sizing supply or borrowing, preparing repay/withdraw calls, or reconciling a V4 execution plan through official Aave MCP or AaveKit. V3 Pools, eMode and Aptos require their own interfaces.
| File | Kind | Size |
|---|---|---|
| ATTRIBUTION.md | markdown | 946 B |
| LICENSE | other | 1.0 KB |
| SKILL.md | skill md | 4.9 KB |
| agents/openai.yaml | config | 236 B |
| references/examples.md | markdown | 1.9 KB |
| references/operations.md | markdown | 3.3 KB |
SKILL.md
View on GitHub---
name: galleon-aave-v4
description: Use when inspecting an Aave V4 Hub/Spoke position, sizing supply or borrowing, preparing repay/withdraw calls, or reconciling a V4 execution plan through official Aave MCP or AaveKit. V3 Pools, eMode and Aptos require their own interfaces.
license: MIT
compatibility: Portable instructions for public official Aave MCP or version-compatible AaveKit and read-only EVM tools. No signer included.
metadata:
author: Andrew Wilkinson and Galleon Labs
version: "0.4.6"
---
# Aave V4 positions and execution plans
Produce a decision for an observed V4 deployment and, when requested, a complete unsigned sequence with health preview and confirmation prerequisites. Do not migrate a V3 recipe by renaming its Pool.
## Discover the live deployment
Inspect the current tool schema and selected version. Use `get_chains` and `get_markets` with `version: v4`, the relevant symbols and actual `user`. For supply/borrow, select the current `get_markets` reserve. For withdraw/repay, use the `reserveId` returned by `get_position_items` for the exact owned position being exited or repaid. A same-symbol market reserve on another Spoke is not that position. Keep the selected opaque ID unchanged. Resolve chain, Spoke, Hub, reserve, asset, decimals and position identity from those observations, never a remembered address or hand-built ID. API coverage is not the universe of Aave deployments: retain covered, omitted and unserved chains.
Read user summary/positions, selected reserve details, relevant Hub asset/liquidity and position items. V4 risk and collateral operate within the position/Spoke; a wallet aggregate can hide a constrained position. Preserve principal, accrued interest, collateral enablement, applicable borrowing/health conditions, caps and available liquidity at one recorded state. Indexed USD rounding cannot settle a dust debt question.
Load [official MCP and SDK procedure](references/operations.md) for exact selectors, input units, plan branches and evidence limits. Refresh chain state when an executable action needs a fresh balance or allowance observation. Do not infer production maturity or deployed chain coverage from SDK `next` naming alone.
## Preview the precise action
Establish action, exact main-unit amount, recipient/on-behalf-of account, selected reserve and user risk floor. Aave MCP amounts are ordinary positive decimal strings in human token units; base-unit integer strings belong to different interfaces. Keep the unit conversion explicit. Inspect optional `native`, `max` and collateral behavior using current schemas.
Call `preview_action` with the exact parameters intended for `prepare_action`. Read returned `data`, all warnings and the health factor/position changes. An error-level warning stops the requested action; a warning needs explanation, not concealment. A health preview does not check every allowance or prove EVM execution. Apply user risk limits and separate interest/oracle stress from current conditions. Plain supply does not provide borrowing collateral unless the chosen action enables it.
## Build and reconcile the sequence
Use `prepare_action` with unchanged selectors and parameters. Inspect `__typename`; return every approval, permit, pre-contract gateway transaction and economic action in provider order. A typed permit creates spend authority and must bind the actual spender, amount, chain/domain and deadline. Review returned `from`, `to`, `data`, `value`, `chainId` and `operations` before a wallet handoff. An unknown plan branch is unresolved.
For separately authorized signing/submission, preserve exact payload and valid existing limits. After onchain approval, wait for the receipt and observed allowance before rebuilding. For V4 operations, carry the builder's `operations` into `get_transaction_processed` with its transaction hash; poll within bounded limits before preparing a dependent next action. This tool is not an arbitrary hash explorer. A provider index flag is not chain finality.
Repay uses current principal/interest and actual spendable debt token. After repay, re-read residual debt and collateral conditions before withdrawal. `withdrawableIgnoringDebt` omits borrowing constraints. Do not automatically increase the repay amount, alter collateral settings or substitute assets beyond user limits.
## Deliver
Return chain/Spoke/Hub/reserve/account, recorded observation state, amounts and units, current/proposed/stressed position, complete unsigned plan, warnings, approval and indexing prerequisites, and next missing read. For completed authorized work, require successful chain receipt plus matching position and actual asset balances; compare the preview with the observed outcome and explain indexing/accrual gaps.
Read [worked records and recovery](references/examples.md). Research checked 2026-10-08; no real supply, borrow, repay or liquidation was performed by this package. This skill is independently installable.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |