Conciseness

The body is dense and signal-heavy — threshold tables, compact Python functions, and an output template with almost no filler. It is not a 5 because passages like 'Perpetual futures have no expiry date. Instead, a funding rate is exchanged between longs and shorts... to keep the perpetual price anchored to the spot price' and the generic 'buy spot + sell perpetual futures → collect funding rate' explanation restate knowledge Claude already has; not a 3 because the padding is minor and most tokens carry information Claude does not have (exchange schedules, thresholds, OKX endpoints).

Actionability

Mostly executable guidance: `funding_regime`, `annualized_basis`, `carry_trade_pnl`, and `oi_funding_matrix` are valid runnable Python, and the carry-trade execution steps name specific instruments ('Buy spot BTC-USDT on OKX spot market', 'short BTC-USDT-SWAP'). It is not a 5 because the OKX API access section shows endpoints only as comments ('# GET /api/v5/public/funding-rate-history?instId=BTC-USDT-SWAP') without a usable request example, and the historical pattern statistics ('70% probability of a 5-10% correction') are asserted without a way to verify or recompute them.

Workflow Clarity

The carry-trade section has a clear numbered 5-step sequence with an exit condition ('Close both legs when funding rate turns negative or basis compresses'), but the overall analysis workflow — fetch rates, classify regime, check basis, assess carry, produce the output — is never laid out as a sequence and must be inferred from the section order and output template. Validation checkpoints are implicit (e.g. no step to sanity-check data freshness or cross-exchange consistency), which matches the 'steps present but checkpoints missing or implicit' anchor rather than the 'most checkpoints present' anchor above.

Progressive Disclosure

No bundle files exist (references/, scripts/, assets/ are absent), and the single external pointer is `load_skill("okx-market")` for data retrieval — referenced but not clearly signaled as a separate navigable resource. At ~250 lines, substantial content that could live in reference files (signal frameworks, threshold tables, output template) is inlined in SKILL.md; sections are well-organized, which lifts it above a 2, but the monolithic structure without split references fits the 'could be better organized; content that should be separate is inline' anchor.

Specificity

The description lists multiple concrete, specific capabilities — 'funding rate analysis and cash-carry basis trading', 'funding rate regimes, annualized basis signals, carry trade construction, and funding rate arbitrage between exchanges' — which map directly to the skill's body sections and comprehensively cover its scope. It is not a 4 because there are no meaningful gaps in coverage relative to what the skill does; the actions are named concretely rather than generically.

Completeness

The 'what' is clear and detailed (regimes, basis signals, carry construction, cross-exchange arbitrage), but there is no 'Use when...' clause or equivalent trigger guidance anywhere in the description. Per the judging guidelines, a missing 'Use when' caps completeness at 3 even with a clear 'what'.

Trigger Term Quality

Good natural keyword coverage: 'perpetual futures', 'funding rate', 'basis', 'cash-carry', 'arbitrage' are phrases a user analyzing perps would actually say. It is not a 5 because common synonyms like 'perp', 'funding rates', or 'basis trade' are missing; it is not a 3 because coverage clearly goes beyond a couple of generic terms and hits the domain's real vocabulary.

Distinctiveness Conflict Risk

This is a clear niche — perpetual funding rates and cash-carry basis trading — with distinct trigger terms (funding rate, basis, carry, perp) that would not plausibly fire for unrelated skills. It is not a 4 because the vocabulary is domain-unique enough that even closely related crypto skills would not be confused with it.