Content
88%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 highly actionable, well-sequenced RE workflow with concrete tool calls, validation checkpoints, and a completion gate. It is efficient and assumes Claude's competence, with only minor trim and structure-split opportunities.
Suggestions
Move the crypto constant reference table and the inline capa/pefile entropy Python scripts into a references/ file (e.g. crypto-constants.md) and link to it, keeping the body leaner.
Tighten the per-question strategies so they reference the core 2a-2e loop rather than restating its steps.
Consider extracting the Tools table into a separate reference file or condensing it, since the same tool calls are already demonstrated inline.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Mostly lean and assumes Claude's RE knowledge without explaining basics, but the crypto constant table and inline capa/entropy Python scripts are somewhat long and the per-question strategies lightly repeat the core loop; minor trim opportunities remain. | 4 / 5 |
Actionability | Fully executable guidance throughout: concrete Ghidra MCP calls with realistic arguments (ghidra_batch_rename, ghidra_retype, ghidra_set_comment) and runnable shell/Python for crypto-constant grepping, capa, and pefile entropy scanning. | 5 / 5 |
Workflow Clarity | Clear numbered loop (2a-2e) with an explicit 'On-Task Check (every 3-5 tool calls)' validation checkpoint, context-rot prevention rules, re-read feedback loop, and a concrete completion gate checklist. | 5 / 5 |
Progressive Disclosure | No bundle files exist, so this scores against the in-body structure, which is well-organized into clearly headed sections; the crypto constant table, inline entropy script, and tools table could arguably be split into reference files, leaving minor organization gaps. | 4 / 5 |
Total | 18 / 20 Passed |