Content
96%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 an exceptionally lean, actionable playbook with executable PoCs, a misconfig matrix, and a validating Decision Gate that sequenced detection through exploitation. Its only minor weakness is that it is a single ~100-line file where a couple of sections (e.g., per-variant PoCs, chains) could optionally be split into references.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The playbook is dense and assumes Claude's competence: a three-line operational opener (severity + framability rule) followed by tight header-inspection commands, a misconfig matrix, and minimal PoCs — every section earns its place with no generic concept padding. | 5 / 5 |
Actionability | Provides copy-paste-ready, fully executable artifacts: a curl header-inspection loop, a saved-to-disk frame probe, and complete HTML PoCs for overlay, drag-and-drop, and double-clickjacking covering the common cases. | 5 / 5 |
Workflow Clarity | A clearly numbered sequence (Detection → Misconfig → PoC → Chains → Tools → Signatures/OPSEC) culminates in a Decision Gate checklist with explicit validation ('If all checked, escalate... otherwise downgrade'), satisfying the destructive-action validation requirement. | 5 / 5 |
Progressive Disclosure | Well-organized into clearly labeled numbered sections in a single file with no nested or buried references and no bundle files; at ~100 lines it is slightly beyond the under-50-line simple-skill exception, so a small organization penalty applies despite the clean structure. | 4 / 5 |
Total | 19 / 20 Passed |