Skip to content

timwhitez/AutoRE-CLI / skills/auto-re

auto-re

Verified

Analyze a supplied executable, DLL, binary or object to explain what it does, investigate suspicious behavior, decompile a function, or inspect strings and references. Use for requests such as 'what does this executable do?' or '这个程序是干什么的', even without naming Auto-RE. Static ELF/PE/Mach-O and explicitly identified raw shellcode only; never run target bytes.

evals/trigger_cases.json

View on GitHub
{
  "schema_version": 1,
  "kind": "auto_re_skill_trigger_cases",
  "cases": [
    {
      "id": "open_ended_pe_triage",
      "prompt": "Statically triage this unknown PE malware sample and give me bounded evidence.",
      "should_trigger": true,
      "expected_route": "report",
      "reason": "Open-ended untrusted binary triage needs the context bundle."
    },
    {
      "id": "known_function",
      "prompt": "Inspect function 0x401000 in this ELF and show only the bounded function evidence.",
      "should_trigger": true,
      "expected_route": "function",
      "reason": "An exact function selector should bypass general triage."
    },
    {
      "id": "known_cfg",
      "prompt": "Show the CFG for function 0x100003f20 in this Mach-O without running it.",
      "should_trigger": true,
      "expected_route": "dump-cfg",
      "reason": "A concrete CFG request has a direct bounded command."
    },
    {
      "id": "pe_strings",
      "prompt": "List the first bounded page of PE strings and their static references.",
      "should_trigger": true,
      "expected_route": "pe-strings",
      "reason": "The user requested one PE inventory surface."
    },
    {
      "id": "go_evidence",
      "prompt": "Inspect this stripped Go binary for bounded Go metadata and function evidence.",
      "should_trigger": true,
      "expected_route": "inspect-go",
      "reason": "The language-specific static inspector is the narrow route."
    },
    {
      "id": "archive_diff",
      "prompt": "Semantically compare these two Auto-RE analysis archives.",
      "should_trigger": true,
      "expected_route": "diff",
      "reason": "Archived analysis comparison is an Auto-RE static workflow."
    },
    {
      "id": "explicit_raw_shellcode",
      "prompt": "Statically inspect this confirmed x86_64 raw shellcode at base 0x1000; do not execute it.",
      "should_trigger": true,
      "expected_route": "report",
      "reason": "Explicit raw identity, architecture, and base permit bounded static triage."
    },
    {
      "id": "sandbox_execution",
      "prompt": "Run this malware sample in a sandbox and record its network traffic.",
      "should_trigger": false,
      "expected_route": null,
      "reason": "Dynamic execution and observation are outside the Skill."
    },
    {
      "id": "debug_process",
      "prompt": "Attach gdb to this process and trace every system call.",
      "should_trigger": false,
      "expected_route": null,
      "reason": "Debugger and runtime tracing requests must not route here."
    },
    {
      "id": "compile_rust_project",
      "prompt": "Compile this Rust workspace for Linux and Windows.",
      "should_trigger": false,
      "expected_route": null,
      "reason": "Ordinary source compilation is not reverse engineering."
    },
    {
      "id": "source_review",
      "prompt": "Review this Rust source pull request for API design issues.",
      "should_trigger": false,
      "expected_route": null,
      "reason": "Source review does not require binary static analysis."
    },
    {
      "id": "elf_concept",
      "prompt": "Explain what an ELF program header is in general.",
      "should_trigger": false,
      "expected_route": null,
      "reason": "A conceptual question has no target to inspect."
    },
    {
      "id": "execute_extracted_script",
      "prompt": "Extract the installer script from this package and execute it.",
      "should_trigger": false,
      "expected_route": null,
      "reason": "Target-derived scripts must remain data."
    },
    {
      "id": "plain_executable_question",
      "prompt": "What does this executable do? The input is /samples/unknown.exe.",
      "should_trigger": true,
      "expected_route": "report",
      "reason": "A concrete binary question needs evidence without requiring the tool name."
    },
    {
      "id": "chinese_plain_binary",
      "prompt": "这个程序是干什么的?文件在 /samples/unknown,没有扩展名。",
      "should_trigger": true,
      "expected_route": "report",
      "reason": "Natural-language intent and a concrete input matter more than extensions or jargon."
    },
    {
      "id": "chinese_suspicious_dll",
      "prompt": "帮我看看 /samples/update.dll 是否有可疑行为,不要运行它。",
      "should_trigger": true,
      "expected_route": "report",
      "reason": "Collect bounded static evidence before making behavioral claims."
    },
    {
      "id": "chinese_known_function",
      "prompt": "看看 /samples/app.exe 里 0x401230 这个函数的逻辑。",
      "should_trigger": true,
      "expected_route": "function",
      "reason": "An explicit function address must not be replaced by generic triage."
    },
    {
      "id": "no_tool_request",
      "prompt": "Do not run any tools. Only explain which commands could inspect /samples/app.exe.",
      "should_trigger": false,
      "expected_route": null,
      "reason": "Respect explicit no-tool instructions even when an input path is present."
    },
    {
      "id": "chinese_concept",
      "prompt": "解释一下反编译和反汇编的区别,不需要分析具体文件。",
      "should_trigger": false,
      "expected_route": null,
      "reason": "Conceptual questions do not require binary inspection."
    },
    {
      "id": "chinese_dynamic_trace",
      "prompt": "运行 /samples/app.exe,抓取它实际发出的网络请求。",
      "should_trigger": false,
      "expected_route": null,
      "reason": "A supplied binary does not bring dynamic execution into scope."
    }
  ]
}
slack-gif-creatoranthropics/skillsKnowledge and utilities for creating animated GIFs optimized for Slack. Provides constraints, validation tools, and animation concepts. Use when users request animated GIFs for Slack like "make me a GIF of X doing Y for Slack."Stars: 198kUpdated Claude CodeCodexVerifiedInstall
ponytailDietrichGebert/ponytailLazy senior dev mode for any coding task: the smallest change that fully solves it, and a reply a busy human understands in one read.Stars: 194kUpdated Claude CodeCodexVerifiedInstall
ponytail-debtDietrichGebert/ponytailHarvest every shortcut: comment into one debt ledger, so deferrals get tracked instead of forgotten. One-shot report.Stars: 194kUpdated Claude CodeCodexVerifiedInstall
ponytail-debtDietrichGebert/ponytailList every `shortcut:` comment (and older `ponytail:` ones) as a debt ledger. One-shot report, changes nothing. Use for "ponytail debt", "what did ponytail defer", "list the shortcuts", /ponytail-debt.Stars: 194kUpdated Claude CodeCodexVerifiedInstall