Conciseness

The body is efficient: rule thresholds, risk/exposure tables, and a CLI command, with no explanation of concepts Claude already knows. Minor trimmable content keeps it below the top anchor — the bilingual aside "rather than '直近 N 取引日' to avoid ambiguity", the free-tier note ("Free tier (250 calls/day) is sufficient"), and implementation details like "ensure_ascii=False" padding the Outputs section. Not score 3 because these are minor instances rather than unnecessary explanations of whole sections.

Actionability

Fully executable, copy-paste-ready CLI command with all flags ("python3 skills/ibd-distribution-day-monitor/scripts/ibd_monitor.py --symbols QQQ,SPY --lookback-days 80 --instrument TQQQ ..."), exact detection thresholds, API-key resolution priority order ("--api-key, config.data.api_key, or FMP_API_KEY env var"), and concrete output file patterns. This matches the anchor for fully executable commands covering the common cases.

Workflow Clarity

The 9-step workflow is clearly sequenced and includes validation checkpoints ("Validate data quality; record skipped sessions in audit", "flag market_below_21ema_or_50ma (None if data insufficient)"). It falls short of the top anchor because there is no explicit error-recovery feedback loop — e.g., what to do when the FMP API call fails or data validation rejects sessions. Not score 3 because checkpoints are present, only the retry/fix guidance is missing.

Progressive Disclosure

The body is well-sectioned and the CLI command points at the real scripts/ bundle, but the two reference files in references/ (ibd_distribution_methodology.md, tqqq_exposure_policy.md) are never mentioned or linked anywhere in SKILL.md, and their content (detection rules, exposure policy) is partially duplicated inline. Per the guideline to score against the actual bundle structure, this fits 'references present but not clearly signaled' — the good_overall pattern of well-signaled one-level-deep references is not met, but structure itself is more than minimal.

Specificity

The description lists multiple concrete actions with exact parameters: "close down at least 0.2% on higher volume", "track 25-session expiration and 5% invalidation", "count d5/d15/d25 clusters", "classify market risk (NORMAL/CAUTION/HIGH/SEVERE)", and "emit TQQQ/QQQ exposure recommendations". Coverage is comprehensive with precise numeric thresholds, matching the anchor for multiple specific concrete actions; it exceeds the 'several specific actions; minor gaps' anchor.

Completeness

It explicitly answers both what (detect, track, count, classify, emit recommendations) and when ("Use after market close, before TQQQ exposure changes, or as input to FTD/market-state frameworks"), plus a boundary statement ("Does not execute trades"). This matches the top anchor: concrete what AND when with explicit trigger phrases, not merely the weaker 'when could be more specific' anchor.

Trigger Term Quality

Strong natural keywords for this domain: "Distribution Days", "QQQ/SPY", "TQQQ exposure", "after market close", "market risk", "FTD/market-state frameworks". A few natural terms are missing — no synonyms like "institutional selling", "Follow-Through Day" spelled out, or "volume decline" — so it does not reach the comprehensive-with-synonyms anchor but is clearly above 'a few natural terms missing' being the only gap.

Distinctiveness Conflict Risk

The IBD distribution-day framing with QQQ/SPY/TQQQ specifics carves a clear niche, but the description shares "classify market risk" / "market-state frameworks" territory with closely related skills (ftd-detector, market-top-detector referenced in the body), creating minor overlap risk. It is mostly distinct rather than having minimal conflict risk, fitting the anchor just below the top.