smartworldarafath/rea_skill

rea-grilling

Grill the user relentlessly about a reverse engineering, lost source code recovery, binary triage, web/mobile security audit, or asset extraction plan before writing code.

查看源码

使用方法

  1. 复制安装命令,在自己的智能体环境安装前先检查源码包。
  2. 从原始文档核对所需工具、凭据和支持的宿主。
  3. 说明输入材料与期望交付物,并对照来源检查输出。
npx skills add https://github.com/smartworldarafath/rea_skill --skill rea-grilling

开始前需了解

  • 安装不包含模型订阅、API 额度或访问文件的权限。
  • 仓库星数和安装量反映活跃情况,不等于独立质量或安全测试。
仓库原始内容

按源仓库内容呈现,保留标题、案例、代码、表格、链接以及原文引用的演示图片。

REA Grilling: Interactive Target & Scope Interrogation Engine

Interview the user relentlessly until you reach a crystal-clear, shared understanding of the target, constraints, and success criteria. Map the investigation as a decision tree: every engineering decision branches into lower-level technical choices.

Work the tree in rounds. The frontier is every decision whose prerequisites are already settled: the questions you can ask now without guessing at answers you haven't heard yet. Ask the whole frontier in one round: number each question and give your recommended answer. Then wait for the user's answers before moving to the next round.

Format each round like this:

markdown
❓ **Q1** - **<question title>**: <question body with clear context and available choices>

➡️ **Recommended**: <your recommended technical answer based on initial artifact inspection>

---

❓ **Q2** - **<question title>**: <question body with clear context and available choices>

➡️ **Recommended**: <your recommended technical answer based on initial artifact inspection>

Crucial Rules:

  1. The "Yes" Rule: Word each question and recommendation so that if the user simply replies "yes", it accepts all of your recommended answers for that round.
  2. Finding Facts is Your Job: Never ask the user for facts you can determine yourself using tools (readelf, DIE, strings, open_binary, inspect_pe_resources, grep, file). Inspect the target first, discover the compilers/architecture/frameworks, and then grill the user only on intent, fidelity, and scope trade-offs.
  3. No Skeletons Guarantee: Once the grilling interview concludes and all frontier branches are settled, compile the final agreed specification and immediately execute the 5-Stage REA Pipeline under the Zero-Skeleton Law.

The REA Decision Tree Frontier

When interrogating a reverse engineering or recovery target, compute questions across these frontier branches:

Branch 1: Scope & Recovery Fidelity

  • Full Idiomatic Rewrite vs. Byte-Exact Decompilation: Do you require 100% byte-exact binary matching, or clean, modern, idiomatic source code (e.g., C++20 / Rust / Go) that compiles cleanly and reproduces the exact feature behavior?
  • Feature Prioritization: Are all UI modules, background workers, and protocols required, or is there a specific core subsystem to reconstruct first?

Branch 2: Toolchain & Compiler Targeting

  • Target Compiler: Which compiler toolchain should the 1-click build script support?
  • Options: Zig CC (zig cc), MSVC (cl.exe), MinGW GCC, Clang, Cargo, Gradle, Go.
  • Dependency Strategy: Should third-party libraries (e.g., SQLite, zlib, OpenSSL, ImGui) be vendor-bundled, dynamically linked, or fetched via package managers (vcpkg, Conan, Cargo)?

Branch 3: Resource & Asset Extraction Depth

  • Embedded Assets: Should binary resources be extracted as loose editable files (valid .ico, .png, .wav, .glsl, .ttf) or linked directly into a compiled .rc resource file?
  • Database & Storage Schemas: Should SQLite/Realm schemas be emitted as raw SQL migration files (schema.sql) or mapped to ORM / typed data models?

Branch 4: Security & Sensitive Artifact Handling

  • Secret Sanitization: When analyzing decompiled code or web bundles, how should extracted API keys, tokens, or credentials be handled?
  • Options: Auto-sanitize into .env.example templates, replace with mock test fixtures, or quarantine securely.
  • Vulnerability Remediation: Should the output include direct inline code patches (remediations) alongside the vulnerability report?

Branch 5: Verification & Acceptance Gates

  • Acceptance Criteria: What constitutes proof of 100% successful recovery?
  • Options: 0 compilation errors and 0 warnings via build.bat / build.sh, successful automated unit/smoke tests, dynamic GUI launch test, or diff coverage against original binary exports.

Interview Execution Flow

  1. Initial Artifact Triage (Silent Fact-Finding):
  • Run REA MCP or local inspection tools to determine: file format, architecture (x86/x64/ARM64), packing/obfuscation, compiler signature, and symbol status.
  1. Round 1 (Target Scope & Fidelity):
  • Present findings from fact-finding.
  • Ask Round 1 frontier questions (Scope, Language/Compiler target, Fidelity level) with your recommended answers.
  1. Round 2 (Toolchains, Resources & Edge Cases):
  • Unblock dependent questions based on Round 1 answers.
  • Address dependencies, asset extraction, database reconstruction, and secret handling.
  1. Round 3 (Verification & Final Handoff):
  • Finalize build scripts, test harness requirements, and acceptance criteria.
  1. Execution Gate:
  • Summarize the settled Recovery Specification.
  • Hand off to $rea-full-code-recovery or the matching domain skill to reconstruct the complete codebase without any skeletons or stubs.
来自同一仓库

更多 Skills

全部 Skills
smartworldarafath
社区

agent-introspection-debugging

Structured self-debugging workflow for AI agent failures using capture, diagnosis, contained recovery, and introspection reports. Use when an agent run fails and you need a reproducible diagnosis instead of a retry.

1小时安装量
2
GitHub Stars
5
最近更新
11月8日
smartworldarafath
社区

agent-reach

Read-only access to internet platforms (Twitter/X, Reddit, YouTube subtitles, GitHub, LinkedIn, Bilibili, web). Use when the agent needs to search, read web links, retrieve social discussions, CVE/vulnerability intel, or when standard web search/fetch is blocked or insufficient.

1小时安装量
2
GitHub Stars
5
最近更新
11月8日
smartworldarafath
社区

agent-sort

Build an evidence-backed ECC install plan for a specific repo by sorting skills, commands, rules, hooks, and extras into DAILY vs LIBRARY buckets using parallel repo-aware review passes. Use when ECC should be trimmed to what a project actually needs instead of loading the full bundle.

1小时安装量
2
GitHub Stars
5
最近更新
11月8日
smartworldarafath
社区

anti-ai-slop-writing

Produces human-sounding text that avoids detectable AI writing patterns. Activates on any writing task — tweets, emails, articles, bios, captions, reports, copy, messages, LinkedIn posts, cover letters, README files, or any content where the output must not read as AI-generated. Enforces banned vocabulary, structural variety, punctuation discipline, accuracy rules, and voice calibration. Use when the user says "write," "draft," "rewrite," "make this sound human," "anti-slop," "not AI," or any va

1小时安装量
2
GitHub Stars
5
最近更新
11月8日