Content
61%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.
A practical, command-driven CLI skill that is mostly executable and well-structured. Its weaknesses are unnecessary marketing prose, broken 'Step 2' cross-references, and missing validation guidance for destructive proxy operations.
Suggestions
Remove the marketing intro paragraph ('Leadoku is a lead generation...') and the auth-plumbing rationale; Claude only needs the commands.
Label the workflow steps explicitly (Step 1: Connect, Step 2: Search actions, Step 3: Run) so the 'skip to Step 2' references resolve, and add a validation/dry-run step before destructive proxy calls (e.g. GET before DELETE/PUT).
Fix the 'Official docs' URL, which currently points to HubSpot (developers.hubspot.com) rather than Leadoku.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The bulk is tight command examples and flag tables, but the opening marketing paragraph ('Leadoku is a lead generation and sales automation platform...') and rhetorical phrases like 'so you can focus on the integration logic rather than auth plumbing' are unnecessary padding Claude does not need. | 3 / 5 |
Actionability | It provides concrete, copy-paste-ready commands with real flags for install, auth, connect, search, run, and proxy, plus a flag reference table; minor gaps remain (only two example actions, and placeholders like CONNECTION_ID/<actionId> require substitution). | 4 / 5 |
Workflow Clarity | A reasonable state-machine sequence with polling feedback loops exists, but the 'skip to Step 2' references are unlabelled, and the proxy path supports destructive verbs (DELETE/PUT) with no explicit verification step, which caps this dimension at 3. | 3 / 5 |
Progressive Disclosure | The single-file body is well-organized into clearly headed sections (Install, Authentication, Connecting, Searching, Running, Proxy, Best practices); it is not a 5 because no content is split into one-level-deep reference files and the file runs ~150 lines. | 4 / 5 |
Total | 14 / 20 Passed |