Content
81%Weight 40%Scale 1-5Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The body is a strong, actionable workflow: concrete commands throughout, a clear five-step sequence with setup validation and error recovery, and well-signaled reference files. Its main weakness is redundancy — the troubleshooting/error table appears twice and the output-schema detail duplicates references/schema.md, which inflates token cost without adding guidance.
Suggestions
Merge the 'Common setup issues' table (Step 1) and the 'Error Reference' table (end) into one location — they repeat 'Extension not connected', 'No session', and 'CSRF token missing' with the same fixes.
Trim the 'Output columns' section to a one-line summary ('see references/schema.md for full column definitions per command') since that reference file already documents the same details.
Reword the host-specific `!` backtick probe block at the top into a plain instruction (e.g., 'run this shell command and use its output') so it is unambiguous regardless of host tooling.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The 'Common setup issues' table in Step 1 and the 'Error Reference' table at the end repeat the same three errors verbatim, and the 'Output columns' section duplicates material that belongs to references/schema.md. It is mostly efficient with no explanations of concepts Claude already knows, so it sits at 'mostly efficient but could be tightened' rather than 'noticeably verbose'. | 3 / 5 |
Actionability | Every step is backed by copy-paste-ready commands ('npm install -g @jackwener/opencli', 'opencli twitter timeline -f json --limit 20', 'opencli twitter search "$AAPL earnings" --filter live --limit 10 -f json'), a user-request-to-command routing table, and concrete flag/format tables covering the common cases — fully executable. | 5 / 5 |
Workflow Clarity | Steps 1–5 are clearly sequenced with an explicit setup probe and decision branches (READY / SETUP_NEEDED / NOT_INSTALLED), `opencli doctor` as a validation checkpoint, a dedicated diagnostics step, and an error-recovery table; the skill is read-only so the destructive/batch validation cap does not apply. | 5 / 5 |
Progressive Disclosure | Two real, one-level-deep reference files are clearly listed with descriptions of their contents and when to read them, and the body serves as an overview. However, the inlined output-columns detail and the duplicated troubleshooting/error tables are minor organization gaps that keep it below the 'content appropriately split' anchor. | 4 / 5 |
Total | 17 / 20 Passed |