paritytech

Build and smoke-test the @parity/truapi TypeScript package (tsc + bun test). Use after regenerating the client or after touching js/packages/truapi/.

Boot the local TrUAPI dev stack (dotli host + playground) in a tmux `servers` window so the playground can be tested in a real browser. Auto-trigger on "run the playground locally", "test locally", "let me try it", "open the playground locally", "boot the stack", or any equivalent phrasing in this repo.

Write an RFC document based on user-provided notes and context. Use when the user wants to create, draft, or write an RFC.

Regenerate the @parity/truapi TypeScript client and playground metadata from the truapi crate's rustdoc JSON. Use whenever the Rust trait surface changes.

The full local end-to-end checklist before declaring a TrUAPI change done. Chains the layered skills in order. Invoke when the user says "is this ready", "definition of done", or asks to verify a Rust→codegen→TS→playground change end-to-end.

Run the playground end-to-end inside the dotli host iframe, either via the automated Playwright suite or by driving it manually in a browser. Use to verify the wire protocol after Rust/codegen/transport changes.

Static verification of the playground (Next.js build + ESLint). Use after changing playground/ sources or after refreshing the @parity/truapi snapshot.

Force-refresh the playground's frozen snapshot of @parity/truapi after the package has been rebuilt. Use whenever js/packages/truapi/ changes (codegen, transport, package.json).

Use when connecting to Polkadot chains, querying chain state, subscribing to storage, working with typed APIs, or choosing between preset and BYOD paths. Covers @parity/product-sdk-chain-client and @parity/product-sdk-descriptors.

Use when uploading or retrieving data via Cloud Storage, working with CID-based decentralized storage, or the CloudStorageClient SDK. Covers the store() upload builder, chunked uploads, container-only reads (fetch/query), CID helpers, and authorization.

End-to-end scaffolding and implementation of Polkadot applications using @parity/product-sdk packages. Use when: creating a new Polkadot project, building a dApp, scaffolding chain interactions, choosing which @parity/product-sdk packages to install, or when a user says "build me a Polkadot app". Handles both developer-guided and fully autonomous (non-developer) workflows.

Submit transactions, connect wallets, manage signers, and handle keys in product-sdk. Use when: submitting transactions, integrating Host API signing (Polkadot Desktop/Mobile), managing multi-provider wallet accounts, deriving keys, creating dev signers for testnet, or wiring QR/mobile sign-in for CLIs. Covers @parity/product-sdk-tx (submit/watch), @parity/product-sdk-signer (wallet connection, account management, multi-provider signing), @parity/product-sdk-keys (key derivation, session keys), and @parity/product-sdk-auth (QR/mobile CLI sign-in + product-account session signers).

Use when migrating a product codebase to @parity/product-sdk — detects legacy stacks (polkadot-apps, novasamatech, hand-rolled crypto/IPFS, polkadot-api 1.x) in the target repo, decides which migration areas apply, writes a per-repo migration spec, then hands off to writing-plans.

Use when interacting with smart contracts (PolkaVM/Solidity) on Asset Hub, using ContractManager with cdm.json manifests, createContract for ad-hoc contracts, ContractRuntime creation, or contract type codegen. Covers @parity/product-sdk-contracts.

Use when publishing or subscribing to ephemeral messages on the Polkadot Statement Store. Covers StatementStoreClient lifecycle, two connection modes (host and local), topic/channel creation, ChannelStore last-write-wins semantics, data size limits, and StatementTransport BYOD.

Use when working with Polkadot address encoding, SS58, H160, EVM address conversion, cryptographic encryption (AES, ChaCha, NaCl, HKDF), byte encoding/decoding, token formatting (planck), key-value storage, or structured logging in @parity/product-sdk packages. Covers address, crypto, utils, storage, and logger utilities.

Use BEFORE writing or changing any non-trivial code in `src/` — new code, new/moved/renamed files, a new abstraction, extending behavior, or a refactor that relocates logic. The single entry point that decides what the change is and where it goes before any file path is generated. Skip only for trivial in-place edits within one existing file.

Use before writing any non-trivial utility, helper, algorithm, data transformation, or wrapper that an installed dependency might already provide. Symptoms — about to hand-roll debounce/merge/groupBy/retry, a stream operator, a validator, or a wrapper around a first-party host/SDK package whose API you don't have in context.

Use BEFORE `domain-development`/`feature-development` when the target layer (domain, aggregate, feature, widget, shared) is uncertain — multi-layer changes, new abstractions, promoting code out of a feature, or "where should this live?". Skip when the layer is obvious.

Use when working in `src/domains/`, introducing/extending a domain's public API (services, resources, hooks, repositories, schemas, README), modeling a user action in domain terms, event-storming domain seams, or deciding whether business logic belongs in a domain.

Use when working in `src/features/` or `src/routes/` files hosting features, scaffolding a feature, wiring into `src/bootstrap.ts`, defining/injecting DI extension points (slots, pipelines, transformers, side effects, SDKs), placing cross-feature logic, or deciding whether two flows are one feature.

Use when writing, reviewing, or refactoring React code — components, hooks, data fetching, bundle optimization, re-render performance, or rendering throughput. Skip for non-React TypeScript, backend, or framework configuration.

Use to audit a branch diff or PR against THIS project's architecture and code rules — after implementing a change, before merging, or when asked to review a PR/branch. Project-rule-aware (cites docs/claude checklists with blocking/major/minor severity); complements the generic /code-review. Reads a plan in docs/_plans/ if present to diff intent vs implementation.

Interact with smart contracts (Solidity and ink!) on Polkadot Asset Hub. Use when: reading contract state, submitting contract transactions, integrating CDM-managed contracts via cdm.json, or working with raw contract ABIs. Covers @polkadot-apps/contracts (ContractManager, createContract, codegen).

End-to-end scaffolding and implementation of Polkadot applications using @polkadot-apps packages. Use when: creating a new Polkadot project, building a dApp, scaffolding chain interactions, choosing which @polkadot-apps packages to install, or when a user says "build me a Polkadot app". Handles both developer-guided and fully autonomous (non-developer) workflows.

Connect to Polkadot chains (Asset Hub, Bulletin, Individuality) using typed APIs from @polkadot-apps/chain-client and @polkadot-apps/descriptors.

Submit transactions, connect wallets, manage signers, and handle keys in polkadot-apps. Use when: submitting transactions, connecting browser wallet extensions (Talisman, Polkadot.js, SubWallet), integrating Host API signing (Polkadot Desktop/Mobile), managing multi-provider wallet accounts, deriving keys, or creating dev signers for testnet. Covers @polkadot-apps/tx (submit/watch), @polkadot-apps/signer (wallet connection, account management, multi-provider signing), and @polkadot-apps/keys (key derivation, session keys).

Use when publishing or subscribing to ephemeral messages on the Polkadot Statement Store. Covers StatementStoreClient lifecycle, two connection modes (host and local), topic/channel creation, ChannelStore last-write-wins semantics, data size limits, and StatementTransport BYOD.

Use when uploading or retrieving data on the Polkadot Bulletin Chain, working with CID-based decentralized storage, IPFS gateway access, or the BulletinClient SDK. Covers upload, batch upload, fetch, query, CID computation, and gateway utilities.

Use when working with Polkadot address encoding, SS58, H160, EVM address conversion, cryptographic encryption (AES, ChaCha, NaCl, HKDF), byte encoding/decoding, token formatting (planck), key-value storage, or structured logging in @polkadot-apps packages. Covers address, crypto, utils, storage, and logger utilities.

Guide runtime release process for Bulletin Chain networks

Update the SDK book documentation to reflect changes in the SDK source code

Run all formatting, linting and cleaning checks before committing code

Build the console UI (same as CI)

Review local changes or a pull request (authoritative review criteria)

Use BEFORE writing code on a non-trivial change. Designs the implementation approach, delegates codebase research to the Explore subagent, writes a structured `.claude/PLAN.md`, and presents it via `ExitPlanMode` for the user's approval. Pair with `/implementer` for execution and `/reviewer` for audit.

Generate a well-formatted Git commit message from a description of changes

Use AFTER `/architect` has produced a plan, when writing or modifying Kotlin code. Reads `.claude/PLAN.md`, validates every edit against its scope, delegates example-finding to the `Explore` subagent, and runs a self-review pass against always-on rules before handing off to `/reviewer`.

Audit a branch diff or GitHub PR against the PolkadotApp architecture and code rules. Cites the checklists as primary source of truth; reads `.claude/PLAN.md` (if present) to diff intent vs implementation. Produces a console comment list grouped by theme and tagged blocking/major/minor.

Audit the doc-comments and prose in a change against CONTRIBUTING.md and fix what violates the rules. Use before committing or opening a PR, or when asked to check changes against CONTRIBUTING.md. Style-only, not a code review.

This skill should be used when the user asks to "deploy", "deploy to previewnet", "deploy to paseo", "deploy browse-beta00", or "ship the app" from a browse-style repo. It builds the current repo's app for the target network (argument — `previewnet` default, or `paseo`/`paseo-next-v2`) and deploys it to a DotNS domain (default `browse-beta00.dot`) on the matching Bulletin environment, handling the recurring gotchas: Bulletin storage-pool authorization, the local `bulletin-deploy` module symlink, and reclaiming `app.`/`widget.` subnames left owned by a prior deployer.

Preact performance optimization guidelines for SPAs built with Vite. This skill should be used when writing, reviewing, or refactoring Preact code to ensure optimal performance patterns. Triggers on tasks involving Preact components, data fetching, bundle optimization, or performance improvements.

Read-only review of a GitHub pull request. Use when asked to review a PR, look at a PR, or assess readiness to land. Never push, merge, or modify code intended to keep.

Deploy smart contracts to Polkadot Asset Hub. Triggers: deploy, deployment, paseo, mainnet

Deploy frontend to Bulletin Chain + DotNS. Triggers: deploy frontend, bulletin, dotns, .dot domain, decentralized hosting

dot.li universal resolver - the fourth host. Triggers: dot.li, smoldot, helia, link sharing, client-side resolution

Foundry testing patterns for Solidity contracts. Triggers: test, testing, forge test, coverage, solidity test

Test Solidity contracts with forge test --polkadot against pallet-revive runtime using EVM or PVM backends

Unified methodology for deterministic AI interaction via Diátaxis (information architecture), COSTAR (prompt structure), and advanced reasoning techniques (CoT, Decision Matrices, Few-Shot). Transform LLMs from text predictors into cognitive decision engines.

Use when creating, refactoring, or validating OpenCode skills. Provides S.T.A.R. workflow, Bottom-Up Reconstruction, and Quality Gates. Semantic cohesion: Contextual granularity with dynamic context pruning.

Run all formatting, linting and cleaning checks before committing code

Start all five UIs (landing + drive-ui + provider-dashboard + s3-ui + photos) locally with hot reload and print the ports table. Use when the user says "run locally", "run the UIs", "start the UIs", "spin up the UIs", or similar — covers any time the goal is to drive the user-interfaces apps in a browser, including verifying changes to the shared network-picker, the landing page, or any individual UI.

Review local changes or a pull request (authoritative review criteria)

Use when fetching or decoding chain data (blocks, extrinsics, events, storage, runtime APIs) and tempted to make manual RPC calls (legacy_rpc, rpc_params!, chain_getBlock, state_call), or when writing code that branches on metadata versions

Use when decoding SCALE bytes, defining value types for dynamic storage queries, or choosing between parity_scale_codec::Decode and scale_decode::DecodeAsType — including when decoded values look subtly wrong without any error

Use when extrinsics unexpectedly report is_signed() == false or a missing signature, or when handling V5 "General" extrinsics/transactions

Provides FRAME pallet development best practices, patterns, and antipatterns for Substrate runtime development. Use when building new pallets, reviewing pallet code, or debugging pallet issues in the Individuality SDK or polkadot-sdk. Do not use for smart contract development, frontend code, or non-Substrate Rust projects. Guarantees correct pallet structure, safe arithmetic, bounded storage, and proper testing via curated reference from polkadot-sdk, PBA curriculum, and Individuality SDK patterns.

Creates and validates skill artifacts with FCIS architecture. Use when the user requests to create, fix, validate, or challenge skills. Do not use for general file editing or installing skill systems. Guarantees FCIS-compliant skills via G1-G9 gate validation and destructive review.