Content
78%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-organized offensive Wi-Fi reference with executable commands across the full attack lifecycle. Its main gaps are progressive disclosure (all content inline, no bundle files) and the lack of an explicit validate-retry feedback loop in the engagement cheatsheet.
Suggestions
Move peripheral sections (Sidebands & Adjacent Wireless, Wi-Fi 6/6E/7 Considerations, Detection / Defender View, RADIUS Backend Pivots) into separate reference files under references/ linked from SKILL.md, keeping the body as an overview + core workflow.
Add an explicit validate->fix->retry feedback loop to the Engagement Cheatsheet (e.g. 'Verify handshake with hcxpcapngtool; if no EAPOL frames, re-deauth and re-capture' before cracking).
Trim minor explanatory asides that restate what the command does, since the commands are self-explanatory to a competent operator.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is command-dense and mostly assumes domain competence, but contains minor explanatory asides ('Triggers heavy crypto on AP CPU; can DoS lower-end deployments', 'Most modern APs lock out after a few failures — slow and noisy') that could be trimmed; length reflects breadth of attack classes covered, not padding. | 4 / 5 |
Actionability | Provides copy-paste-ready, executable commands throughout (airmon-ng, airodump-ng, aireplay-ng, hcxpcapngtool, hashcat, eaphammer, reaver, hostapd-mana, mdk4) plus concrete tool tables covering the common cases. | 5 / 5 |
Workflow Clarity | A 'Quick Workflow' (5 steps) and a numbered 'Engagement Cheatsheet' (7 steps) give a clear sequence with some verification moments ('Verify monitor + injection', 'Verify the EAPOL frames are usable', 'Targeted capture if PMKID empty'), but the main cheatsheet lacks an explicit validate->fix->retry feedback loop. Validation is present rather than missing, so the destructive/batch cap does not apply; it sits just below 5. | 4 / 5 |
Progressive Disclosure | The body is well-structured with clear section headers and tables, but everything is inlined into a single ~340-line SKILL.md with no separate reference files (none of references/, scripts/, assets/ exist), so peripheral topics that could be split out are inline. Good section organization prevents a 2; absence of any one-level-deep reference split keeps it at 3. | 3 / 5 |
Total | 16 / 20 Passed |