Flare Builders Toolkit

Flare Builders Toolkit — Master Skill

This skill is the entry point to a 16-skill bundle for Flare-family EVM development. When invoked, route to the right sub-skill(s) based on what the task needs.

Security is the default. Every contract under discussion is assumed hostile-attacked. Patterns that aren't safe-by-default (Ownable instead of Ownable2Step, raw transfer instead of SafeERC20, unbounded loops, push-style ETH transfers, missing reentrancy guards) are flagged as findings, not described as alternatives. Read flare-security first for any contract design or review work.

How to use this skill

When the user asks a Flare-development question, route them to the most specific sub-skill that fits. Multiple sub-skills can apply to one question. Don't try to answer from this skill alone — it's the router, not the source of truth.

Routing table

Standard flow for a new Flare contract

When the task is "build and ship a Flare contract", walk these in order. Read the linked skills as you go — don't try to remember everything.

1. Plan — read flare-general plus any protocol-specific skills the contract will touch (flare-ftso, flare-fdc, flare-fassets, flare-smart-accounts, enosys-contracts, enosys-dex-v3).
2. Address resolution — every concrete address goes through flare-network. Don't guess; don't trust training-data addresses; look them up.
3. Coding standards — solidity Cyfrin standards as the baseline.
4. Flare-specific security overlays — flare-security covers what solidity doesn't: Permit2 chain availability, FoT detection, blacklist surface, basefee floors, FTSO redistributor proxy quirks. Add flare-dapp-pitfalls for the gotcha catalog.
5. Test — test-foundry for unit + fork + fuzz coverage. (Hardhat alternative in test-hardhat; Foundry is preferred.)
6. Optimize gas — gas-optimize for ranked savings.
7. Audit — audit (systematic, 115+ items) + audit-contract (adversarial, multi-agent). After both run, reread flare-security Part 2 — Flare-specific overlays aren't in the generic audit checklists.
8. Deploy + verify — verification commands in flare-network work for both Blockscout (flare-explorer) and Routescan/Etherscan-style (flarescan).

Pre-deploy checklist (mandatory)

Before broadcasting to Flare mainnet, every item below must be green. Full detail in flare-security.

• forge build clean; all contracts under 24 KB runtime
• forge test (unit) + forge test --fork-url flare (fork) all pass
• Slither no high/medium findings
• Owner is NOT the deployer EOA — Ownable2Step + Ledger/multisig handoff
• Ownable2Step (not Ownable); reentrancy guards on every state-changing external function; CEI in every flow
• SafeERC20 + forceApprove for every token interaction
• No tx.origin for auth; no unbounded loops over user-controlled data
• Pause / emergency-stop present and tested
• Custom errors prefixed with contract name + __
• Permit2 chain availability verified if the contract depends on Permit2
• FoT and blacklistable-token surface explicitly designed for
• Basefee floor present in any incentive-math
• audit + audit-contract both run; findings fixed or accepted
• Contract verifies cleanly on both flare-explorer and flarescan

Conventions

• Sub-skill folders are flat. Each contains a SKILL.md (frontmatter + body) plus optional references/, scripts/, or *.md supporting docs.
• Protocol skills (flare-ftso, flare-fassets) include reference scripts in TypeScript and Solidity. They're illustrative; the user is responsible for security review before signing or broadcasting.
• Address registry (flare-network) is the single source of truth for any deployed-contract or token address. If a sub-skill mentions a different address than flare-network, raise it as a discrepancy and verify.

When NOT to use this skill

• For non-EVM chain interactions outside the FAssets / FDC bridge surface — use chain-specific tooling.
• For Flare validator-node operation or staking management — see Flare's validator docs.
• For executing transactions — this skill is documentation/reference only. Wallet signing happens in user-controlled environments.

Contract under review?

If the user is asking you to review a specific contract, start with flare-security to set the security baseline, then run audit and audit-contract against the file. After those complete, reread flare-security Part 2 to verify the Flare-specific overlays — Permit2 availability, FoT detection, blacklist surface, basefee floor, FTSO redistributor handling — were caught. The generic audit skills are not chain-aware.

For more detailed orientation, the README.md in this same directory is a human-readable map of the toolkit.