Content
76%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 highly actionable with executable examples across JS/TS, Python, and REST, and is well-structured and concise. The main gap is the absence of explicit validation/retry feedback loops for batch Actor runs, which caps workflow clarity.
Suggestions
Add an explicit validation/retry feedback loop for Actor runs: after a run finishes with a non-SUCCEEDED status, show how to inspect logs, adjust input, and re-run before consuming dataset results.
Consider extracting the REST API reference and per-language deep dives into separate reference files (e.g., rest-api.md, python-async.md) and linking to them from SKILL.md to improve progressive disclosure.
Tighten the Best Practices section into terser bullet guidance to lift conciseness toward fully lean.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Dense, code-heavy body with almost no padding of concepts Claude already knows (no 'what is Apify' prose); a few minor trims are possible in the Best Practices prose, placing it at the 'efficient; minor over-explanation' anchor rather than fully lean (5) or padded (3). | 4 / 5 |
Actionability | Provides copy-paste ready install commands and complete executable TS/Python snippets with imports and token setup, plus full REST calls with headers and payloads, covering the common cases (sync, async, retrieval, errors). | 5 / 5 |
Workflow Clarity | Sections are sequenced conceptually (token, find actor, install, execute, retrieve, errors, best practices), but batch operations (running Actors, paginating large datasets) lack an explicit validate->fix->retry feedback loop; per the rubric's destructive/batch cap, workflow clarity is capped at 3 even though the sequence is present. | 3 / 5 |
Progressive Disclosure | Well-organized sections with clearly signaled one-level-deep external doc links; no bundle files exist, so all content is inlined in a ~205-line file where substantial reference material (full REST examples, per-language deep dives) could arguably be split into separate files, leaving minor organization gaps rather than a fully split structure. | 4 / 5 |
Total | 16 / 20 Passed |