Content
63%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 research-orchestration skill with concrete queries, fallbacks, and useful domain heuristics, weakened by a monolithic single-file layout: the long report template is inlined (and its scorecard duplicated) instead of being split into a one-level-deep reference file. Splitting the template and LinkedIn-scrape details into references/ and trimming duplication would lift both conciseness and progressive disclosure.
Suggestions
Move the ~150-line report template to references/report-template.md and reference it from Phase 4, keeping SKILL.md as an overview.
Deduplicate the 15-dimension scorecard, which appears both in section 3A and inside the output template.
Extract the LinkedIn scraping procedure (2B) into a reference file and note clearly that the scraper script lives in the external linkedin-profile-post-scraper skill.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The instruction sections are mostly efficient and the hiring-signal heuristics ("First sales hires = transitioning from PLG to sales-assisted") are genuine domain knowledge, but the ~150-line report template is fully inlined and the 15-dimension scorecard appears twice (section 3A and again inside the output template). Mostly efficient but could be tightened — the 3 anchor; not 4, since the duplication and inlined template are more than minor trimmable instances. | 3 / 5 |
Actionability | Concrete, near copy-paste-ready guidance throughout: specific WebSearch/WebFetch query patterns ("site:<website> vs OR alternative OR compare"), fallback paths for JS-rendered sites, named job-board URL patterns, and an explicit scraper command with flags. Not 5 because the LinkedIn scraper command depends on a script in an external skill (skills/linkedin-profile-post-scraper) not present in this bundle, leaving a minor execution gap. | 4 / 5 |
Workflow Clarity | Clear Phase 1–4 sequence with explicit parallelization guidance, per-thread goals, and fallback checkpoints, plus a "Don't guess — flag unknowns" ([NEEDS VERIFICATION]) instruction. Matches the 4 anchor (clear sequence, most checkpoints present); not 5, since there is no explicit validate-and-retry loop confirming collected data before synthesis. | 4 / 5 |
Progressive Disclosure | Good section structure and headers, but there are no bundle files at all and the entire research protocol plus the full report template are inlined in a ~480-line SKILL.md. Content that clearly belongs in a separate reference file (the report skeleton, the LinkedIn scraping procedure) is inline — the 3 anchor; not 4, since the split into references/ is absent rather than a minor organization gap. | 3 / 5 |
Total | 14 / 20 Passed |