Content
55%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 executable examples and clear workflows, but suffers severely from verbosity and lack of progressive disclosure. Information is repeated across Common Actions, Workflow, Commands Reference, Examples, and Tips sections, inflating the token cost significantly. The content would benefit greatly from splitting the exhaustive command reference and examples into separate files, keeping SKILL.md as a concise overview with the workflow and most common patterns.
Suggestions
Deduplicate aggressively: the same commands and concepts (default social set, --use-default, scratchpad, LinkedIn mentions, multi-platform posting) appear in 3-5 different sections. Pick one canonical location for each and remove the rest.
Move the Commands Reference tables and the extensive Examples section into separate files (e.g., COMMANDS.md, EXAMPLES.md) and reference them from SKILL.md to dramatically reduce the main file size.
Consolidate the Tips section into the relevant workflow/command sections rather than repeating information already covered elsewhere in the document.
Remove explanatory text that Claude already knows (e.g., what social media platforms are, what rate limits mean) and trust Claude to infer context from the command syntax.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is extremely verbose at ~450+ lines. There is massive redundancy: the same information about default social sets, --use-default flags, and platform options is repeated across multiple sections (Common Actions, Workflow, Commands Reference, Examples, Tips). The Examples section alone contains 30+ examples that largely duplicate the Commands Reference table. Character limits, scratchpad usage, and LinkedIn mentions are each explained in multiple places. Claude doesn't need this level of repetition. | 1 / 3 |
Actionability | Every command is fully executable with exact CLI syntax, flags, and expected outputs. The examples are copy-paste ready with concrete commands like `./scripts/typefully.js drafts:create --text "Hello, world!"`. The Common Actions table maps user intents directly to specific commands. | 3 / 3 |
Workflow Clarity | The workflow section provides a clear numbered sequence for creating posts, starting with checking config, resolving social sets, creating drafts, and scheduling/publishing. The API key error handling section has explicit validation steps (tell user, stop and wait, do not attempt alternatives). The multi-platform workflow has a clear create-then-update sequence. Publishing confirmation acts as a safety checkpoint for destructive operations. | 3 / 3 |
Progressive Disclosure | The entire skill is a monolithic document with no references to supporting files despite being extremely long. The Commands Reference tables, extensive Examples section, and detailed LinkedIn mentions guide could all be split into separate reference files. No bundle files are provided, and the content that should be in separate files (API reference, examples catalog) is all inline, creating a wall of text. | 1 / 3 |
Total | 8 / 12 Passed |