Conciseness

The body is efficient — mostly executable commands with brief, substantive context lines and no explanation of concepts Claude already knows. Not 5 because example commands hardcode time-sensitive dates (e.g. '--as-of 2026-06-28', '--from 2020-01-01 --to 2026-06-28') outside any deprecated/old-patterns section, and the 'Shared FMP client' section includes implementation details (key masking, retry counts) that could be trimmed.

Actionability

All five subcommands (scan, enrich, update-outcomes, summarize, backfill) are given as complete, copy-paste-ready invocations, and every flag shown (e.g. '--fmp-universe', '--prices-json', '--survivorship-complete', '--horizons', '--group-by') exists in the bundled run_20pct_study.py CLI. Offline and FMP modes plus expected outputs are both covered.

Workflow Clarity

Steps 1-5 are clearly sequenced with command-level detail and several checkpoints: best-effort enrichment degradation to 'NO_CLEAR_NEWS', the conditional '--survivorship-complete' guardrail, '--min-sample 10', and the requirement of 'out-of-sample validation before changing trade rules'. Not 5 because there is no explicit verify-output/re-run loop after the batch scan and backfill steps (e.g. checking event counts or flagged records before proceeding).

Progressive Disclosure

A Resources section clearly signals one-level-deep references that map exactly to real bundle files (all five references/*.md plus scripts/run_20pct_study.py and scripts/fmp_client.py), with inline purpose descriptions. Not 5 because the bundled assets/cohort_summary_template.md and assets/daily_report_template.md are not referenced or navigated from SKILL.md.

Specificity

The description lists four concrete actions — 'scanning +20%/-20% movers, classifying catalysts and setup context, updating forward outcomes, and summarizing cohort patterns' — plus backfilling, matching the comprehensive multi-action anchor. Not 4 because coverage is essentially complete across the skill's actual workflow stages rather than having identifiable gaps.

Completeness

It explicitly answers both 'what' (build and maintain a daily 20% mover study via scanning, classification, outcome updates, cohort summaries) and 'when' ('Use when the user asks to run a daily 20% study, backfill historical 20% movers, find recurring edge patterns, or build a model book'), matching the anchor with concrete trigger phrases.

Trigger Term Quality

Trigger phrases like 'run a daily 20% study', 'backfill historical 20% movers', 'find recurring edge patterns', and 'build a model book of explosive market moves' are natural for this niche, and 'Stockbee-style' is a distinctive keyword. Not 5 because common user phrasings such as 'which stocks moved 20% today', 'top/big gainers', or '20 percent' variants are missing.

Distinctiveness Conflict Risk

'Stockbee-style daily 20% mover study for US equities' carves out a clear niche with distinct, specific triggers that would not plausibly fire for unrelated skills; the research/model-book framing further separates it from generic market-scan skills.