Conciseness

The body is mostly dense, useful signal — a command routing table, flag tables, and concrete examples — with no explaining of concepts Claude already knows. It falls short of the lean anchor because of trimmable redundancy: the "Common setup issues" and "Error Reference" tables repeat the same three rows (Extension not connected / No session / CSRF token missing), `opencli doctor` is explained three times (Step 1, setup, Step 5 Diagnostics), and the read-only caveat is restated three times in the body.

Actionability

Guidance is fully executable and copy-paste ready: a routing table mapping each user request to a specific command with its flags ("opencli twitter search \"QUERY\"" with "--filter top|live", "--limit N"), worked examples ("opencli twitter search \"$AAPL earnings\" --filter live --limit 10 -f json"), and per-command output columns. It matches the top anchor — concrete commands covering the common cases — and exceeds the 4 anchor which allows minor gaps.

Workflow Clarity

Steps 1–5 form a clear sequence with a real checkpoint (run `opencli doctor` before anything else if unsure; jump to Step 5 on failure). The gap versus the 5 anchor is that the Step 1 status check is a non-executable pseudo-block — "!`(command -v opencli && opencli doctor ...) || echo \"NOT_INSTALLED\"`" — with an unexplained `!`(...) syntax, and recovery guidance is duplicated across three sections rather than a single feedback loop. It sits above the 3 anchor since checkpoints are explicit, not merely implied.

Progressive Disclosure

Structure is good: SKILL.md is a working overview, both referenced files (references/commands.md, references/schema.md) exist, are one level deep, contain no further nested references, and are clearly signaled ("Read the reference files when you need exact command syntax..."). It misses the top anchor because the detailed "Output columns" section (per-command column lists) duplicates material that belongs in references/schema.md, keeping some reference content inline.

Specificity

The description enumerates many concrete capabilities — "the user's home timeline, tweet search, trending topics, bookmarks, a user's recent tweets, threads, articles, profiles, followers and following, and notifications" — plus an explicit boundary ("cannot post, like, retweet, reply, or follow"), matching the comprehensive-coverage anchor. It exceeds the 4 anchor ('several specific actions; minor gaps') because coverage of read operations is essentially complete.

Completeness

It clearly answers 'what' (reads timeline, search, trending, bookmarks, profiles, etc. through opencli) and 'when' with an explicit trigger clause: "Use this skill whenever the user wants to know what people are saying on Twitter/X or fintwit about a stock, topic, or market event". This matches the top anchor ('both what AND when with concrete trigger phrases') and exceeds the 4 anchor where 'when' could be more specific.

Trigger Term Quality

Natural user phrasing is well covered: "what people are saying on Twitter/X or fintwit about a stock, topic, or market event", "check their feed or bookmarks", "see what's trending", "look up an @account and its posts", with the synonym "fintwit" and both "Twitter/X" forms. These are phrases a user would naturally say, matching the comprehensive-synonyms anchor rather than the 4 anchor where a few natural terms are missing.

Distinctiveness Conflict Risk

It carves a clear niche — read-only Twitter/X access for financial research via opencli — with distinct triggers (fintwit, market events, @account lookups) and an explicit "Read-only: it cannot post, like, retweet, reply, or follow" boundary that prevents overlap with any posting/social-management skill. Minimal conflict risk, matching the 'clear niche with distinct triggers' anchor.