Everything here is original prose and original snippets, written against public documentation and open-source SDK code. The initial API verification was on 2026-08-16; later checks and their limits are recorded below.
| Source | Used for | Licence |
|---|---|---|
| Hyperliquid docs | every endpoint, action, field, limit and error string in the hyperliquid-* skills |
public docs |
| hyperliquid-python-sdk 0.24.0 | Python calls in the skills; the SDK the desk installs | MIT |
| @nktkas/hyperliquid 0.33.3 | TypeScript calls and formatting helpers | MIT |
| decimal.js 10.6.0 | Exact decimal arithmetic in the TypeScript bound example; install separately | MIT |
| Grok Bot docs, Cursor help | Bots, group chats, shared computer, skills, routines, approvals, secrets | public docs |
| Grok Build skills and plugins, Agent Skills spec | plugin layout and SKILL.md conventions |
public docs |
| Senpi skills, cezar-r/hyperliquid-skills, Hermes hyperliquid skill | structure survey only; nothing reused | MIT / Apache-2.0 declarations |
| kaileycompact51/HyperLiquid-Claw | reviewed and not used; it distributes an unverified Windows binary and an obfuscated install script | claimed MIT |
Order price/size precision was rechecked against the official tick and lot size contract on 2026-09-21. scripts/test_order_rounding.py executes only the published pure Decimal helpers, including buy-ceiling and sell-floor cases. No signed order was submitted for this validation.
The fork review records all observed heads and dispositions. Ideas from thysmans/hypergrok-iterated, jawndiego/hypergrok-trading-desk and shadowslayerai/hypergrok-trading-desk were adapted into original prose; no third-party code, hooks or skill bodies were copied. Existing MIT notices remain intact.
Account abstraction and userRole response fields, subscription acknowledgements, snapshot fills and fill identifiers were rechecked against the official info endpoint and WebSocket subscriptions references on 2026-10-02. Reservations and reconnect readiness are desk controls, not exchange guarantees. No key or signed order was used for this review.
The idea of a tool-layer confirmation came from thysmans/hypergrok-iterated (see the fork review). hooks/exchange_write_gate.py is an original stdlib implementation written against the Claude Code hooks and plugin manifest references: ask forces a prompt even in auto mode, and a plugin hooks path merges with any default hooks file. Method names come from the SDK calls already documented in hyperliquid-orders, hyperliquid-positions and hyperliquid-advanced. No fork code was copied.
Reviewed hypurrclaw/hyperliquid-cli, version 0.11.1, at commit 5d950fff8b8c1fcd9792132614fec103476f666f. Its command contracts and explicit DEX resolution informed the original stdlib JSON diagnostics, offline contract discovery and read-only HIP-3 snapshots in this release. No third-party code, prose, fixture, wallet backend, fee default or execution loop was copied.
The review exposed two existing inconsistencies: float-first spot formatting and an incorrect claim that a newer nonce invalidates all lower nonces. Corrections were checked against the official tick and lot rules, nonce model, noop action and info endpoint. Decimal values are preserved through the published helpers and checked against Python SDK 0.24.0's float_to_wire before the signing boundary. Offline regression tests exercise approved price bounds, structured failures, source identity, delisted/ambiguous markets and explicit DEX requests. Live proof uses public /info reads only; no key or signed action is used.
The original research collector and offline replay verifier were designed against the official info endpoint, perpetuals metadata and history, funding rules and rate limits. The references establish the recent-candle ceiling, canonical market identity, history pagination, hourly funding and oracle-based payment arithmetic. No external collector, fixture or strategy implementation was copied.
Research datasets distinguishes exchange behavior from local controls: explicit closed UTC ranges, strict data validation, finite page budgets, duplicate conflict handling, saved decoded responses, coverage gaps and offline replay. Hour-bucket funding coverage preserves actual millisecond event times. Historical research can retain a currently delisted market; current metadata is not a historical universe. Complete coverage does not establish source authenticity, exact funding PnL, an unbiased experiment or Risk PASS.
Release 1.6.0 was tagged and published to npm on 2026-10-12. Its manifest hashes do not prove native skill installation, public template export or marketplace acceptance; the 1.6.0 template export is pending. The 1.5.0 native publication record in Grok Bot template is retained as evidence for that earlier release.