Content
85%Reviews 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 with a clear multi-step workflow and validation gating, only slightly weakened by verbosity in the conceptual and best-practices sections.
Suggestions
Trim the 'Key Insight' paragraph and obvious 'Best Practices' items to reduce token load on concepts Claude already knows.
Consider moving the detailed 'Why coverage, not per-holder reconciliation' rationale into a reference file, keeping only the grade definition inline.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is mostly efficient and procedural, but padded sections like the 'Key Insight' paragraph on what institutions manage, 'Best Practices' truisms, and the lengthy reconciliation rationale add tokens Claude largely already knows. | 2 / 3 |
Actionability | Provides fully executable, copy-paste-ready commands with concrete flags and thresholds (e.g. 'python3 scripts/track_institutional_flow.py --top 50 --min-change-percent 10') plus specific interpretation tiers. | 3 / 3 |
Workflow Clarity | Steps 1-5 are clearly sequenced, and the A/B/C data-reliability grade with automatic exclusion of Grade C stocks serves as an explicit validation/gating checkpoint between screening and interpretation. | 3 / 3 |
Progressive Disclosure | SKILL.md is an organized overview pointing to three real, one-level-deep reference files and real scripts, each clearly signaled by path, with content appropriately split between workflow and deep guides. | 3 / 3 |
Total | 11 / 12 Passed |