Content
53%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 delivers strong, phase-by-phase executable guidance with concrete tooling for every kill-chain stage, but it is undermined by an orphaned bundle: six provided files are never referenced, meaning the report template, engagement tracker scripts, and ATT&CK standards reference are invisible to a reader. Mid-engagement validation checkpoints are also missing for a workflow whose operations (credential dumping, lateral movement, exfiltration) carry high operational risk.
Suggestions
Add a navigation section (e.g., under References) pointing to the bundle files: "Report template: see assets/template.md", "ATT&CK technique listings: see references/standards.md", "Engagement lifecycle detail: see references/workflows.md", "Engagement tracker: see scripts/process.py", "ATT&CK API usage: see references/api-reference.md and scripts/agent.py" — and move the 13-row ATT&CK tactic table into references/standards.md, which already contains it.
Insert validation checkpoints between phases for the risky operations: e.g., verify C2 callback and opsec hygiene before post-exploitation, run deconfliction checks with the SOC liaison before lateral movement, and re-confirm scope/RoE constraints before each objective, with explicit "if detected, execute emergency stop per RoE" handling.
Trim the overview paragraph (red team vs. pentest distinction is knowledge Claude already has) and rewrite the circular "When to Use" bullets to be concrete; replace the nonexistent proxyshell_exploit.py reference with a real, obtainable command or an explicit note that the operator supplies the exploit.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Mostly efficient command blocks, but three areas are padded: the overview paragraph ("Unlike penetration testing, red team operations prioritize stealth, persistence...") explains concepts Claude already knows, the "When to Use" bullet "When conducting security assessments that involve conducting full scope red team engagement" is circular filler, and the 13-row ATT&CK tactic table duplicates both Claude's knowledge and references/standards.md. | 3 / 5 |
Actionability | Concrete, mostly copy-paste-ready commands across all phases (amass, theHarvester, msfvenom, nmap, Rubeus, secretsdump, dnscat2), matching the "mostly executable with minor gaps" anchor: `python3 proxyshell_exploit.py` references a script that does not exist in the bundle, and the GoPhish step is only a comment ("Configure SMTP profile, email template with pretexted lure") with no actual configuration guidance. | 4 / 5 |
Workflow Clarity | The six engagement phases are clearly sequenced with prerequisites acting as an authorization gate, but mid-engagement validation checkpoints (opsec verification, SOC deconfliction, stop conditions) are absent or only implied via a single "emergency stop procedures" mention in Prerequisites; risky operations such as credential dumping and lateral movement lack verification steps, which caps this dimension at 3. | 3 / 5 |
Progressive Disclosure | The body is well-sectioned, but none of the six bundle files (references/api-reference.md, references/standards.md, references/workflows.md, scripts/agent.py, scripts/process.py, assets/template.md) are referenced anywhere in the body — the References section lists only external URLs — so the report template, tracker scripts, and standards listing are undiscoverable, and the inlined ATT&CK table is content that belongs in references/standards.md. Not 3 because references are entirely absent rather than merely unclear; not 1 because the body itself is navigable with clear section headers. | 2 / 5 |
Total | 12 / 20 Passed |