Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add instructions/paritytech/polkadot-apps/claude-mdgit clone --depth 1 https://github.com/paritytech/polkadot-appsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/paritytech/polkadot-apps/claude-md)<a href="https://agentmods.dev/instructions/paritytech/polkadot-apps/claude-md"><img src="https://agentmods.dev/badge/instructions/paritytech/polkadot-apps/claude-md.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.02168 | $0.02168 |
| Opus 5 | $0.01084 | $0.01084 |
| Sonnet 5 | $0.00434 | $0.00434 |
| Haiku 4.5 | $0.00217 | $0.00217 |
Grade A, and why
polkadot-apps CLAUDE.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 5d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 110 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
Follow the contributor guidelines in README.md.
Quick Reference
- Build:
pnpm build - Dev:
pnpm dev - Clean:
pnpm clean - Test:
pnpm test - Test with coverage:
pnpm test:coverage - Format:
pnpm format - Format check:
pnpm format:check - Generate descriptors:
pnpm generate-descriptors - Generate docs:
pnpm docs
Key Conventions
- All packages live in
packages/<name>/and are scoped under@polkadot-apps/. - Use in-source testing (
if (import.meta.vitest)blocks) for unit tests. Separatetests/*.test.tsfiles are for integration tests only. - Internal deps use
"workspace:*", shared versions use"catalog:"frompnpm-workspace.yaml. - Packages must be framework-agnostic pure TypeScript. No React/Vue imports in core packages.
- Biome handles formatting — do not use prettier or eslint.
- Changesets are required for any change to published packages. Run
pnpm changesetand include the generated file in your commit.
Descriptors
@polkadot-apps/descriptorsprovides per-chain subpath imports (e.g.,@polkadot-apps/descriptors/bulletin). Each subpath is a separate papi build — no barrel import, no unnecessary chain metadata.- Each chain lives in
packages/descriptors/chains/<name>/with its own.papi/polkadot-api.jsonconfig.pnpm generate-descriptorsfetches metadata and runspapi generateper chain. - Adding a new chain: (1) add
papi addinscripts/generate.sh, (2) createchains/<name>/with.papi/polkadot-api.jsonandpackage.json, (3) add chain toCHAINSinscripts/build.sh, (4) add subpath export inpackage.json.
Chain Client
@polkadot-apps/chain-clientoffers two ways to connect:- BYOD (
createChainClient): bring your own descriptors and RPCs. Zero overhead from unused chains. - Preset (
getChainAPI): zero-config with built-in descriptors and RPCs for known environments (paseo, polkadot, kusama).
- BYOD (
- Both return a
ChainClient<T>with typed APIs per chain key,.rawforPolkadotClientaccess, and.destroy(). @polkadot-api/sdk-inkis not a dependency of chain-client. Consumers who need contracts createInkSdkthemselves:createInkSdk(client.raw.assetHub, { atBest: true }).isInsideContainerSync()is available from both@polkadot-apps/hostand@polkadot-apps/chain-clientfor synchronous container detection in performance-critical code paths. The asyncisInsideContainer()(which uses product-sdk) remains the primary API.- Bulletin RPC endpoints are imported from
@polkadot-apps/host's shared chain config (BULLETIN_RPCS), not hardcoded in presets. - Preset RPC endpoint lists in
packages/chain-client/src/presets.tsare ordered live-first.@polkadot-api/ws-providerwalks the array on connect and on unclean reconnects, so healthy providers belong at the top and recently-degraded ones at the bottom as fallbacks. Each dead entry at the top costs one "Unable to connect" log line per cold start; rotate rather than remove when a provider is transiently unhealthy so failover redundancy is preserved when it recovers.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 5d ago First seen · 110 lines · 2,168 tokens per session scan A 14003df4a0c0
polkadot-apps CLAUDE.md is an instructions file published in the GitHub repository paritytech/polkadot-apps (2 stars, last pushed 26d ago), licensed Apache-2.0. It adds 2,168 tokens to every session, about $0.0108 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other instructions, from other repositories
tevm CLAUDE.md
Claude Code instructions for evmts/tevm, covering tevm monorepo commands and style guide, project overview, forking implementation, json-rpc support and commands.
starknet-agentic CLAUDE.md
Claude Code instructions for keep-starknet-strange/starknet-agentic, covering starknet agentic -- development context, cairo, typescript, skills and git.
solidity-typescript-hardhat-template CLAUDE.md
Claude Code instructions for protokol/solidity-typescript-hardhat-template, covering claude.md — project instructions, project overview, key commands, solidity conventions and test conventions.
sui-pilot CLAUDE.md
Claude Code instructions for contract-hero/sui-pilot, covering doc-source routing, usage and keeping docs up to date.
agent-sdk CLAUDE.md
Claude Code instructions for trustless-ai/agent-sdk, covering agent-sdk, what this repo is, adding or updating erc support, repo layout and recompute-to-verify.
ponscli CLAUDE.md
Claude Code instructions for mesutgulecen/ponscli, covering ponscli, the rule that matters most, layout, verifying a change and things that will bite.