Now liveThe Skillselion MCP - thousands of ranked skills, loaded into your agent mid-task. No install.Get it →
vechain avatar

Vechain Core

  • 73 installs
  • 9 repo stars
  • Updated June 11, 2026
  • vechain/vechain-ai-skills

Helps with ai & agent building tasks.

About

vechain-core is a Claude Code skill for ai & agent building. It helps solo builders move faster with AI-assisted coding.

  • vechain-core
  • AI & Agent Building
  • AI-coding skill

Vechain Core by the numbers

  • 73 all-time installs (skills.sh)
  • +2 installs in the week ending Aug 5, 2026 (Skillselion tracking)
  • Ranked #5,587 of 16,544 AI & Agent Building skills by installs in the Skillselion catalog
  • Data as of Aug 5, 2026 (Skillselion catalog sync)
npx skills add https://github.com/vechain/vechain-ai-skills --skill vechain-core

Add your badge

Show developers this skill is listed on Skillselion. Paste this into your README.

Listed on Skillselion
Installs73
repo stars9
Last updatedJune 11, 2026
Repositoryvechain/vechain-ai-skills

What it does

Helps with ai & agent building tasks.

Files

SKILL.mdMarkdownGitHub ↗

VeChain Core Skill

CRITICAL RULES

1. Read reference files FIRST. When the user's request involves any topic in the reference table below, read those files before doing anything else — before writing code, before making decisions. Briefly mention which files you are reading so the user can confirm the skill is active (e.g., "Reading fee delegation reference..."). 2. Information priority for VeChain topics: (a) Reference files in this skill — always the primary source. (b) VeChain MCP tools — use @vechain/mcp-server for on-chain data, transaction building, and live network queries; use Kapa AI MCP for VeChain documentation lookups. (c) Web search — only as a last resort, and only for topics NOT covered in the reference files. 3. Prefer working directly in the main conversation for VeChain tasks. Plan mode and subagents do not inherit skill context and may fall back to web search instead of using reference files. 4. After compaction or context loss, re-read this SKILL.md to restore awareness of the reference table and operating procedure before continuing work.

Scope

Use this Skill for general VeChain development:

  • SDK usage (@vechain/sdk-core, @vechain/sdk-network, ethers adapter)
  • Fee delegation (VIP-191) — gasless transactions, backend sponsorship, vechain.energy
  • Multi-clause transactions — atomic batching of multiple operations
  • Dual-token model (VET for value, VTHO for gas)
  • Legacy migration from Connex/thor-devkit to VeChain SDK
  • General VeChainThor development patterns and reference links

For specialized topics, see the companion skills:

  • frontend — Generic frontend patterns: React Query, Turborepo, state management, Chakra UI, i18n, transaction UX
  • vechain-kit — VeChain Kit and dapp-kit packages: hooks, components, wallet connection, social login
  • smart-contract-development — Solidity, Hardhat, testing, security, gas optimization
  • vebetterdao — X2Earn apps, B3TR/VOT3, governance, VeVote
  • stargate — NFT staking, validators, delegation, VTHO rewards

Default stack

LayerDefaultAlternative
SDK@vechain/sdk-core + @vechain/sdk-network@vechain/sdk-ethers-adapter
NodeNode 20 LTS (managed via nvm)--

Operating procedure

1. Check Node version

Before installing dependencies or running any command:

  • Check if .nvmrc exists in the project root. If yes, run nvm use to switch to the required version.
  • If .nvmrc does not exist, create one with 20 (Node 20 LTS) and run nvm use.

2. Detect project structure

  • turbo.json present → follow Turborepo conventions (apps/, packages/*)

3. Clarify before implementing

When the user's request is ambiguous or could be solved multiple ways, ask before building. Do not silently research alternatives and pick one. Separate research from implementation:

  • If the scope is unclear, ask the user to narrow it
  • If multiple architectures are viable, present trade-offs and let the user choose
  • Only proceed to implementation once the approach is agreed upon

4. Implement with VeChain-specific correctness

  • Network: always explicit (mainnet/testnet/solo)
  • Gas: estimate first, use fee delegation where appropriate
  • Transactions: use multi-clause when batching benefits atomicity or UX
  • Tokens: VET for value, VTHO for gas (dual-token model)

5. Verify and deliver

A task is not complete until all applicable gates pass:

1. Code compiles — no build errors 2. Tests pass — existing tests still pass; new logic has test coverage 3. Risk notes documented — any signing, fee, or token-transfer implications are called out

Then provide:

  • Files changed + diffs
  • Install/build/test commands
  • Risk notes for signing, fees, token transfers

Reference files

Read the matching files BEFORE doing anything else. See Critical Rules above.

TopicFileRead when user mentions...
Fee delegationreferences/fee-delegation.mdgasless, sponsored, VIP-191, delegator, vechain.energy
Multi-clausereferences/multi-clause-transactions.mdbatch, multi-clause, atomic, multiple operations
Legacy migrationreferences/sdk-migration.mdConnex, thor-devkit, migration, deprecated
Reference linksreferences/resources.mddocs URL, npm link, GitHub repo

Related skills

This week in AI coding

Five minutes, every Monday - the tools, releases and tactics for developers.

unsubscribe anytime.