Content
93%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.
An efficient, highly actionable emulation playbook with a clearly sequenced kill chain, strong safety gates, and clean one-level routing to per-phase skills. The only gap is that validation/abort checkpoints are stated in surrounding sections rather than embedded as explicit steps in the workflow sequence itself.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean CTI-derived guidance — a two-line framing blockquote, a copy-ready JSON seed, an 11-row kill-chain table, and short OPSEC/RoE/fidelity sections — with no explanation of concepts Claude already knows and no padding to trim. | 5 / 5 |
Actionability | Fully concrete guidance: a complete ThreatProfile JSON seed, per-phase routing to exact skill paths, named tools (Sliver, Foundry/Anvil, NetExec, canary wallets), named output files (deconfliction.json, cleanup.json), and a specific EMERGENCY-abort string. As a routing/instruction skill, no code absence penalty applies. | 5 / 5 |
Workflow Clarity | The kill chain is clearly sequenced twice (table rows 1–11 and CONOPS steps 1–5) and risky on-chain operations have safety gates (EMERGENCY abort criterion, canary wallets, RoE), but those checkpoints live in prose rather than being woven into the numbered sequence as explicit verify-before-proceeding steps. | 4 / 5 |
Progressive Disclosure | The body is a well-organized overview that routes detail to separate per-phase skill files via explicit one-level-deep paths, with one clearly signaled external reference; no content is inlined that belongs in a separate file, and the skill bundle itself contains no additional files. | 5 / 5 |
Total | 19 / 20 Passed |