Content
75%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 highly actionable, well-structured skill whose commands all check out against the real script and whose workflow section gives an agent a clear end-to-end sequence. The main costs are token redundancy from duplicated command/output sections and the absence of any reference files to split out detail content.
Suggestions
Merge the "Two Modes" and "Usage" sections — the same scrape and search commands are shown twice; keep one usage section with mode subsections.
Collapse the CSV columns table and the "What You Get Per Person" bullet list into a single output-format reference (or move it to a references/output-format.md file), since they currently duplicate each other.
Cite the Apify actor rent URL once (in Setup) instead of three times across Two Modes, Setup, and Troubleshooting.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Mostly efficient — no explanations of concepts Claude already knows, and the data-access table is genuinely useful — but there is structural redundancy: the same scrape/search commands appear in both "Two Modes" and "Usage", the CSV columns table and "What You Get Per Person" repeat identical information, and the actor rent URL appears three times. This is more than the minor trimming of the 4 anchor but well short of the heavy padding of the 2 anchor. | 3 / 5 |
Actionability | Every usage example is a copy-paste-ready command, all documented flags (--search, --events-only, --output, --json, --no-cache, --cache-hours) match the actual argparse definitions in scripts/scrape_event.py, setup gives exact URLs and export commands, and troubleshooting covers concrete failure modes. This matches the 5 anchor: fully executable commands covering the common cases. | 5 / 5 |
Workflow Clarity | The "AI Agent Workflow" lays out a clear 5-step sequence (find events → extract profiles → qualify → enrich → generate outreach) with example prompts, and the Troubleshooting section provides error-recovery paths for the main failure modes. It is not a 5 because there are no explicit validation checkpoints (e.g., verifying the CSV was produced and has expected rows before qualifying), and not a 3 because the sequence is explicit and recovery guidance is present. | 4 / 5 |
Progressive Disclosure | Good structure: clear section headers, both bundle scripts (scripts/scrape_event.py, scripts/apify_client.py) referenced with real one-level-deep paths, and easy navigation. It falls short of the 5 anchor because there is no reference layer at all — output-format detail, the options reference, and example prompts are all inline in a ~230-line SKILL.md — which is a "minor organization gap" (4) rather than the inlined-API-reference problem of the 3 anchor, since the operational logic is properly externalized in scripts. | 4 / 5 |
Total | 16 / 20 Passed |