Content
73%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 well-structured, highly actionable skill body with an excellent validated workflow and sound progressive disclosure. Its main weakness is token efficiency: duplicated external-resource listings and issue-tracker changelog narration pad the file.
Suggestions
Move the 'FMP transport (rate limiting, single 429 retry... Issue #495)' sentence out of the bash setup code block so the block stays copy-paste executable.
Deduplicate the WhaleWisdom/SEC EDGAR/DataRoma listing (appears in Step 3, the track_institution_portfolio.py section, and Support & Resources) into a single section.
Relocate the 'Why coverage, not per-holder reconciliation' rationale and the 13F filing-schedule table into a reference file to tighten SKILL.md.
Fix the inconsistent script path in Step 3 ('skills/institutional-flow-tracker/scripts/...' vs 'scripts/...' used elsewhere).
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Mostly efficient domain-specific content (13F schedule, signal thresholds, grade criteria), but the WhaleWisdom/SEC EDGAR/DataRoma list is repeated three times and changelog-style prose ('Issue #495', 'replaces the retired /api/v3/institutional-holder feed') is repo-history noise that could be tightened or moved to a reference. | 3 / 5 |
Actionability | Copy-paste-ready commands for the common cases (quick scan, sector scan, custom screening, single-stock deep dive) plus full parameter documentation, but a prose sentence about FMP transport sits inside the bash setup code block and the Step 3 invocation uses an inconsistent script path. | 4 / 5 |
Workflow Clarity | Five clearly sequenced steps with explicit validation: the A/B/C coverage-based reliability grade gates screening results (Grade C excluded) and the signal-strength framework provides an interpretation checklist with explicit feedback loops ('If institutions are exiting, re-evaluate your thesis'). | 5 / 5 |
Progressive Disclosure | Good structure against the actual bundle: three real reference files clearly signaled with descriptions in two places, documented scripts, one level deep. Not 5 because SKILL.md inlines detail that belongs in references, such as the long 'Why coverage, not per-holder reconciliation' rationale. | 4 / 5 |
Total | 16 / 20 Passed |