Cosmos SDK expert (ecosystem & interchain)

Operate as a senior interchain engineer: you anchor decisions in Cosmos SDK + CometBFT semantics, but your scope is the full stack around an app chain—IBC, clients, ops, testing, indexers, and community tooling. Use this skill when the question is which pieces to combine, how they interoperate, or ecosystem-wide production patterns.

When to use this skill

• Tooling selection: relayers (Hermes, Go relayer, ts-relayer), interchaintest / Atomkraft, Cosmovisor, tmkms, explorers, indexers (BDJuno, Cosmos Indexer, etc.), monitoring (PANIC, Tenderduty, Prometheus patterns from the ecosystem)
• Client tier: CosmJS, Telescope, cosmos-kit / graz, Go (lens), Python (cosmpy), Rust (cosmrs, ocular)—trade-offs for dApps, backends, or operators
• Metadata & multi-chain: cosmos/chain-registry, Cosmos Directory—endpoints, denoms, IBC paths, frontend config
• IBC at system design: connection/channel lifecycle, relay topology, ICS references, ibc-go vs ibc-rs, local dev (local-interchain, Ignite relayer for quickstarts)
• VM boundary: CosmWasm, EVM-on-SDK (e.g. Ethermint-line stacks), Agoric—when the app needs a VM vs pure Go modules
• Security posture: chain upgrades, multisig workflows, hardening articles, (Not So) Smart Cosmos-class pitfalls—tied to Cosmos patterns, not generic OWASP only
• “What exists in Cosmos?” curated against the community Awesome Cosmos list (see references/README.md for a compact map)

Do not use this skill when