Content
57%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, well-structured adversary-emulation profile with concrete, capability-mapped emulation guidance, but it is long and monolithic with no progressive disclosure to separate files, and its destructive-operation workflow lacks explicit validation checkpoints.
Suggestions
Move the full TTPs-by-tactic catalog and/or campaign timeline into a references/ file (e.g., TTPS.md, CAMPAIGNS.md) and keep SKILL.md as an overview with one-level-deep links.
Add explicit validation checkpoints to the emulation workflow (e.g., 'verify the test tenant state before pivoting', 'confirm canary data landed in the in-scope bucket before declaring exfil success', 're-validate scope after each phase').
Tighten the intro/attribution prose to the operational essentials an engagement needs, trimming background narrative on group history and member demographics.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Mostly efficient operational content (TTPs, tooling, detection) but the long intro narrative and attribution background ('predominantly native English-speaking cybercriminal collective', member demographics) is explanation that could be tightened, fitting anchor 3 rather than 4. | 3 / 5 |
Actionability | The emulation-guidance section gives concrete, specific direction — named tools, commands (Get-ADUser, nltest, Rclone), and ATT&CK technique IDs mapped to Decepticon capabilities — with only minor gaps from pointing to other skills rather than inlining every step, matching anchor 4. | 4 / 5 |
Workflow Clarity | The emulation guidance is sequenced into clear phases (initial access → identity pivot → AD → lateral/C2 → virtualization → collection/exfil → impact), but it lacks explicit validate→fix→retry checkpoints for destructive operations (ransomware/EDR-disable emulation), so the destructive-operations cap holds at 3. | 3 / 5 |
Progressive Disclosure | Sections are well-organized with clear headers, but the >150-line body is entirely inline with no bundle files or one-level-deep references; large blocks (full TTP catalog, campaign timeline) that could be split into reference files are inlined, fitting anchor 3 rather than 4. | 3 / 5 |
Total | 13 / 20 Passed |