Content
81%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 a dense, highly actionable command reference that respects token budget and points to one well-signaled reference file. Its main limitation is the lack of explicit validation/feedback loops for destructive batch operations, which caps workflow clarity.
Suggestions
Add explicit validation checkpoints to multi-step destructive flows (e.g., after a secretsdump or relay run, verify output is non-empty/well-formed before proceeding, with a fix-and-retry step).
Move the heavier per-technique detail (stealth tiers, detection indicators) into references/lsass-techniques.md and keep SKILL.md as a tighter overview to improve progressive disclosure.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Lean, command-first reference that assumes Claude's competence; the brief intro states purpose without explaining known concepts, and every section earns its tokens with executable commands or tables. | 5 / 5 |
Actionability | Copy-paste-ready commands with concrete flags, a hashcat mode table, and tool/flag tables cover the common cases across all major credential-access techniques. | 5 / 5 |
Workflow Clarity | Sequences and some checkpoints exist (lockout-awareness rules, PetitPotam→ntlmrelayx two-terminal combo, Decision Gate), but this is a destructive/batch skill without explicit validate→fix→retry feedback loops in most command chains, so it is capped at 3. | 3 / 5 |
Progressive Disclosure | Well-organized sections with a single one-level-deep reference (references/lsass-techniques.md) clearly signaled with a descriptive 'Read when...' pointer; minor gap is that most technique detail is inlined in SKILL.md rather than split into the reference. | 4 / 5 |
Total | 17 / 20 Passed |