Content
62%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The skill is highly actionable with excellent workflow clarity, providing concrete commands, specific UI element targets, and robust validation/error-recovery steps throughout. However, it suffers significantly from verbosity and repetition — the same workflows (regular posts, video, quote tweets, articles) are described three times for three execution modes with only minor variations, and mode-selection boilerplate is repeated under every post type. The content would benefit greatly from consolidation to reduce its token footprint by roughly 40-50%.
Suggestions
Consolidate the three execution modes by describing each post-type workflow once with mode-specific variations called out inline (e.g., 'In Chrome Plugin mode, use file chooser; in Computer Use mode, use native file picker; in CDP mode, run x-browser.ts'), rather than repeating entire workflows three times.
Remove the repeated 'Codex mode note' boilerplate from every post type section — the mode selection logic at the top already covers this and doesn't need restating 4+ times.
Factor shared X Article image-insertion steps into a single reusable section referenced by both Chrome Plugin and Computer Use modes, since the steps are nearly identical.
Consider moving the detailed CDP Script Mode post-type sections (Regular Posts, Video Posts, Quote Tweets, X Articles with parameter tables) into a separate reference file, since CDP is explicitly a fallback mode.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is extremely verbose at ~300+ lines with massive repetition. The three execution modes (Codex Chrome Plugin, Chrome Computer Use, CDP Script) each repeat nearly identical workflows for regular posts, video posts, quote tweets, and X Articles. The 'Codex mode note' boilerplate is copy-pasted under every post type section. Mode selection logic is restated multiple times. Much of this could be consolidated. | 1 / 3 |
Actionability | The skill provides fully executable bash commands with concrete parameters, specific UI element references (aria-labels, menu paths like 'Insert -> Media'), exact error messages to watch for, and detailed step-by-step procedures. Code examples are copy-paste ready with clear parameter tables. | 3 / 3 |
Workflow Clarity | Multi-step workflows are clearly sequenced with explicit validation checkpoints (e.g., verify placeholder count is 0, post-composition check for remaining placeholders, preview verification before publish). Error recovery is well-defined (retry logic for Chrome plugin failures, kill CDP instances on port conflicts). The publish safety gate requiring explicit user confirmation is consistently enforced across all modes. | 3 / 3 |
Progressive Disclosure | The skill references external files (references/regular-posts.md, references/articles.md, EXTEND.md) which is good, but the main SKILL.md itself is a monolithic wall of text with heavily duplicated content across modes that could be better organized. The three mode sections repeat similar workflows that could be factored into shared steps with mode-specific variations called out separately. | 2 / 3 |
Total | 9 / 12 Passed |