复制安装命令
用 Codex 或 Claude 安装复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它先审查 Skill 页面再帮你安装。
复制前请先查看来源、License 和安全提示。
Offensive security skills for Claude — drop-in SKILL.
用 Codex 或 Claude 安装复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它先审查 Skill 页面再帮你安装。
复制前请先查看来源、License 和安全提示。
来源文件:README.md

Offensive security skills for Claude — drop-in SKILL.md files that turn Claude into a context-aware red team operator.
Built by SnailSploit — GenAI Security Research.
claude-red is a curated library of offensive security skills for the Claude Skills system. Each skill is a structured SKILL.md file that primes Claude with expert-level methodology for a specific attack surface — from SQLi to shellcode, EDR evasion to ADCS abuse.
Drop a skill into your Claude environment and it behaves like a specialist: it knows the techniques, the tooling, the edge cases, and the escalation paths. Skills load on demand based on conversational triggers — you don't pay context for skills you aren't using.
Use it for: authorized red team engagements, bug bounty triage, security research, CTF preparation, training operators, and exploring attack surfaces methodically.
# Clone into a directory Claude will scan
git clone https://github.com/SnailSploit/claude-red ~/.claude/skills/claude-red
# Or install only one category
git clone --filter=blob:none --sparse https://github.com/SnailSploit/claude-red
cd claude-red && git sparse-checkout set Skills/web Skills/active-directory
Claude will auto-load matching skills based on conversational triggers (e.g. mentioning SQLi loads offensive-sqli).
# Point Claude at a single skill before a session
cat Skills/web/offensive-sqli/SKILL.md | claude --system-file -
# Or load a whole category
cat Skills/active-directory/**/SKILL.md | claude --system-file -
Paste the contents of a SKILL.md into a Project's system prompt or prepend to your conversation.
./install.sh # interactive
./install.sh --target ~/.claude/skills # explicit target
./install.sh --category web # one category
| Category | Skills | Focus |
|---|---|---|
| Web Application | 16 | OWASP Top 10 + business logic + advanced web bug classes |
| Auth & Identity | 2 | JWT, OAuth |
| Active Directory | 1 | On-prem AD attack methodology (expanding) |
| Wireless | 13 | 802.11, WPA2/3, EAP, WPS, evil-twin, BLE, Zigbee, Z-Wave, LoRa, sub-GHz |
| Cloud | 1 | AWS / Azure / GCP attack paths (expanding) |
| Mobile | 1 | Android + iOS pentest (expanding) |
| IoT & Embedded | 1 | Hardware, firmware, RTOS, ICS (expanding) |
| Infrastructure & Red Team | 7 | Initial access, EDR evasion, Windows ops |
| Exploit Development | 6 | Stack/heap, mitigations, crash analysis, TOCTOU |
| Fuzzing & VR | 4 | libFuzzer, AFL++, bug ID, vuln classes |
| Reconnaissance | 2 | OSINT tooling and methodology |
| AI Security | 1 | Prompt injection, jailbreaks, RAG poisoning |
| Utility | 2 | Fast-checking, professional reporting |
Skills/web/
| Skill | Description |
|---|---|
offensive-sqli | SQL injection — error/blind/OOB, DB-specific, ORM CVEs, cloud paths |
offensive-xss | Cross-site scripting — stored, reflected, DOM, mutation |
offensive-ssrf | Server-side request forgery — cloud metadata, filter bypass |
offensive-ssti | Server-side template injection — engine ID, RCE paths |
offensive-xxe | XML external entity — OOB exfil, blind exploitation |
offensive-idor | Insecure direct object references — enumeration, business logic |
offensive-file-upload | File upload — extension bypass, polyglots, webshells |
offensive-rce | Remote code execution — chaining, command injection |
offensive-deserialization | Insecure deserialization — Java/PHP/.NET gadget chains |
offensive-race-condition | Race conditions — TOCTOU, single-packet, limit bypass |
offensive-request-smuggling | HTTP request smuggling — CL.TE, TE.CL, h2 desync |
offensive-open-redirect | Open redirect — OAuth abuse, phishing, SSRF pivots |
offensive-parameter-pollution | HTTP parameter pollution — WAF bypass, logic confusion |
offensive-graphql | GraphQL — introspection, batching, IDOR via aliases |
offensive-waf-bypass | WAF bypass — encoding, chunking, case mutation |
offensive-business-logic | Business logic — workflow bypass, pricing, refunds, chains |
Skills/auth/
| Skill | Description |
|---|---|
offensive-jwt | JWT — alg:none, key confusion, secret cracking |
offensive-oauth | OAuth — open redirect abuse, token leakage, PKCE bypass |
Skills/active-directory/
| Skill | Description |
|---|---|
offensive-active-directory | AD — Kerberoast, ASREProast, ACL abuse, ADCS ESC1-15, delegation, persistence, hybrid AAD |
Note: This category is being expanded. The AD overview is being split into 16 focused skills (Kerberoasting, ASREProasting, ADCS, coercion, NTLM relay, BloodHound, ticket forgery, GPO abuse, etc.). See Roadmap.
Skills/wireless/
| Skill | Description |
|---|---|
offensive-wifi | 802.11 overview — entrypoint into the wireless category |
offensive-wifi-recon | Adapter selection, monitor mode, multi-band airspace mapping |
offensive-wpa2-psk | Handshake capture, PMKID, hashcat 22000 cracking |
offensive-wpa3-sae | Transition-mode downgrade, Dragonblood, SAE side-channels |
offensive-wpa-enterprise | 802.1X / EAP attacks, eaphammer evil-twin RADIUS |
offensive-wps | Pixie Dust, online PIN brute, vendor PIN generators |
offensive-evil-twin | KARMA, Mana, captive portal, post-association MITM |
offensive-krack-fragattacks | KRACK + FragAttacks supplicant testing |
offensive-deauth-disassoc | Targeted/broadcast deauth, PMF awareness, action frames |
offensive-bluetooth-ble | BLE GATT enum, pairing downgrade, sniffing, MITM |
offensive-bluetooth-classic | BR/EDR — SDP, SPP, KNOB, BlueBorne, HID spoofing |
offensive-zigbee-thread-matter | 802.15.4 mesh — KillerBee, Touchlink abuse, ZCL command injection |
offensive-z-wave | S0 key derivation flaw, S2 commissioning, hub pivots |
offensive-lorawan-sub-ghz | LoRaWAN ABP/OTAA, KeeLoq garage doors, fixed-code, TPMS |
Skills/cloud/
| Skill | Description |
|---|---|
offensive-cloud | AWS / Azure / GCP — privesc, IMDS, cross-account, persistence, CSPM evasion |
Note: Cloud-identity (Entra/AAD/Okta hybrid) skills coming separately. See Roadmap.
Skills/mobile/
| Skill | Description |
|---|---|
offensive-mobile | Android + iOS — Frida, pinning, storage, biometric, deep links |
Skills/iot/
| Skill | Description |
|---|---|
offensive-iot | Hardware recon, firmware, RTOS, ICS/OT, MQTT/CoAP |
Note: Being split into 10 focused skills (UART/JTAG, flash dump, fault injection, U-Boot, secure boot, RTOS, ICS protocols). See Roadmap.
Skills/infrastructure/
| Skill | Description |
|---|---|
offensive-initial-access | Phishing, drive-by, supply chain — TA0001 |
offensive-advanced-redteam | Full kill chain, C2, OPSEC, lateral, persistence |
offensive-edr-evasion | Unhooking, indirect syscalls, PPID spoofing |
offensive-shellcode | Writing, encoding, injection techniques |
offensive-keylogger-arch | Keylogger architecture and input-capture techniques |
offensive-windows-mitigations | Windows mitigations — ACG, Arbitrary Code Guard |
offensive-windows-boundaries | Defeating Windows boundaries — sandbox escape, privilege |
Skills/exploit-dev/
| Skill | Description |
|---|---|
offensive-exploit-development | Stack/heap, ROP chains, mitigations |
offensive-exploit-dev-course | Structured curriculum format |
offensive-basic-exploitation | Linux exploitation, mitigations disabled — beginner-to-mid |
offensive-crash-analysis | Crash triage, exploitability assessment, root cause |
offensive-mitigations | Modern kernel mitigations — ASLR, CFG, CET, PAC |
offensive-toctou | Time-of-check/use across binary, kernel, web, container |
Skills/fuzzing/
| Skill | Description |
|---|---|
offensive-fuzzing | libFuzzer, AFL++, coverage-guided, mutation strategies |
offensive-fuzzing-course | Curriculum — finding vulns via fuzzing |
offensive-bug-identification | Code review patterns, static analysis triggers |
offensive-vuln-classes | Vulnerability classes — real-world examples, taxonomy |
Skills/recon/
| Skill | Description |
|---|---|
offensive-osint | OSINT tools — recon-ng, theHarvester, Maltego pipelines |
offensive-osint-methodology | OSINT methodology — structured intelligence collection |
Skills/ai/
| Skill | Description |
|---|---|
offensive-ai-security | AI pentest — prompt injection, jailbreaking, RAG poisoning |
Skills/utility/
| Skill | Description |
|---|---|
offensive-fast-checking | Fast triage checklist — quick-win identification |
offensive-reporting | Pro pentest reporting — CVSS, evidence, exec summary, retest |
The library is being expanded in seven phases. Track progress in CHANGELOG.md.
| Phase | Category | New Skills | Status |
|---|---|---|---|
| 1 | Internal AD/Windows (rename active-directory/ → internal/) | +16 | Planned |
| 2 | Cloud Identity (Entra/AAD, ADFS, Okta, M365) | +10 | Planned |
| 3 | Wireless split (WPA2/3, EAP, BLE, Zigbee, Z-Wave, LoRa, sub-GHz) | +12 | Mandatory |
| 4 | IoT split (UART/JTAG, flash, fault injection, RTOS, ICS) | +10 | Planned |
| 5 | Web Basics (recon, auth bypass, access control, CSRF, headers, CORS, cache, clickjack) | +8 | Planned |
| 6 | Web Advanced (proto pollution, SAML, OIDC, WebSocket, gRPC, postMessage, SSI/ESI, CSTI) | +10 | Planned |
| 7 | Polish (README, LICENSE, manifest, install) | — | In progress |
End state: ~107 skills across the same 13+ categories.
Contributions welcome. See CONTRIBUTING.md for the skill template, frontmatter standard, and review process. Focused, single-surface skills are preferred over monolithic overviews.
MIT — use freely, attribution appreciated.
"Give Claude the right skill and it stops being a chatbot. It becomes an operator."
This project's full writeup, methodology, and related research lives at:
https://snailsploit.com/claude-red
Created by Kai Aizen — independent offensive security researcher.
snailsploit.com · Research · Frameworks · GitHub · LinkedIn · ResearchGate · X/Twitter
Same attack. Different substrate.
name: offensive-fuzzing
description: "Practical offensive fuzzing methodology covering target identification, fuzzer selection (AFL++, libFuzzer, Honggfuzz, Boofuzz, syzkaller), harness writing, corpus curation, mutation strategies, coverage measurement, and crash triage. Use when setting up or running fuzz campaigns against any target: file parsers, network protocols, kernel drivers, EDR engines, embedded firmware, or language runtimes."| Type | Coverage | Speed | Tools |
|---|---|---|---|
| BlackBox | Poor | Fast | Peach, Boofuzz |
| GreyBox | Good | Fast | AFL++, Honggfuzz, libFuzzer, WinAFL |
| Snapshot | Good | Fastest | Nyx, wtf, Snapchange |
| WhiteBox | Best | Slow | KLEE, QSYM, SymSan |
| Ensemble | Best | Fast | AFL++ + Honggfuzz + libFuzzer |
GreyBox sub-variants: Directed (AFLGo, UAFuzz), Grammar (AFLSmart, Tlspuffin), Concolic (QSYM, Driller), Kernel (syzkaller, kAFL, wtf).
Research target → Choose analyses → Build harness → Seed corpus → Instrument → Fuzz → Triage crashes → Report
copy_from_user — DMA-BUF ops, page fault handlers, VM operation structs, allocation callbacks# AFL++ (preferred for GreyBox)
CC=afl-clang-fast CXX=afl-clang-fast++ cmake -DCMAKE_BUILD_TYPE=Release .. && make -j
# libFuzzer + ASan/UBSan (C/C++)
cmake -DCMAKE_CXX_FLAGS="-fsanitize=fuzzer,address,undefined -O1 -g" ..
# CmpLog build for hard compares
AFL_LLVM_CMPLOG=1 CC=afl-clang-fast CXX=afl-clang-fast++ make clean all
Windows (MSVC): Project Properties → C/C++ → Address Sanitizer: Yes (/fsanitize=address)
libFuzzer (C++):
#include <cstdint>
#include <cstddef>
extern "C" int LLVMFuzzerTestOneInput(const uint8_t* data, size_t size) {
parse_or_process(data, size);
return 0;
}
Honggfuzz HF_ITER (persistent mode — preferred for large targets):
#include "honggfuzz.h"
int main(int argc, char** argv) {
initialize_target(); // runs once
for (;;) {
size_t len; uint8_t *buf;
HF_ITER(&buf, &len);
FILE* s = fmemopen(buf, len, "r");
target_function(s);
fclose(s);
reset_target_state();
}
}
AFL++ persistent mode (__AFL_LOOP):
while (__AFL_LOOP(10000)) {
// re-read input and process
}
macOS IPC (Mach message fuzzing):
void *lib_handle = dlopen("libexample.dylib", RTLD_LAZY);
pFunction = dlsym(lib_handle, "DesiredFunction");
afl-cmin -i raw_corpus -o seeds -- ./target @@afl-tmin -i crash -o crash.min -- ./target @@AFL++ parallel (primary + secondary with cmplog):
afl-fuzz -M f1 -i seeds -o findings -x dict.txt -- ./target @@
afl-fuzz -S s1 -i seeds -o findings -c 0 -- ./target @@
libFuzzer:
./target_libfuzzer corpus/ -max_total_time=3600 -workers=4
Binary-only (QEMU):
afl-fuzz -Q -i seeds -o findings -- target.exe @@
Snapshot (AFL++ Nyx):
NYX_MODE=1 AFL_MAP_SIZE=1048576 afl-fuzz -i seeds -o findings -- ./target_nyx @@
Ensemble (AFL++ + Honggfuzz sharing corpus):
# Terminal 1
afl-fuzz -M fuzzer1 -i seeds -o sync_dir -- ./target @@
# Terminal 2
../honggfuzz/honggfuzz -i sync_dir/fuzzer1/queue -W sync_dir/hfuzz \
--linux_perf_ipt_block -t 10 -- ./target ___FILE___
If progress stalls:
-c 0 on AFL++ secondaries-x dict.txt or AFL_TOKEN_FILEAFL_MAP_SIZE=1048576, -L 0 for MOpt scheduler# 1. Minimize
afl-tmin -i crash -o crash.min -- ./target @@
# 2. Symbolize
ASAN_OPTIONS=abort_on_error=1:symbolize=1 ./target crash.min 2>asan.log
# 3. Hash + bucket
./cov-tool --bbids ./target crash.min > cov.hash
./bucket.py --key "$(cat cov.hash)" --log asan.log --out triage/
Sanitizer env quick reference:
ASAN_OPTIONS=abort_on_error=1:symbolize=1:detect_stack_use_after_return=1
UBSAN_OPTIONS=print_stacktrace=1:halt_on_error=1
TSAN_OPTIONS=halt_on_error=1:history_size=7
MSAN_OPTIONS=poison_in_dtor=1:track_origins=2
| Bug Class | Oracle |
|---|---|
| Memory safety | ASan, HWASan (AArch64, lower overhead) |
| Uninitialized reads | MSan |
| Concurrency | TSan |
| Undefined behavior | UBSan |
| Type safety | TypeSan |
| Heap hardening | Scudo Hardened Allocator |
| Logic bugs | Differential / idempotency oracles |
| Kernel memory | KASAN, KMSAN, KCSAN |
| Kernel UB | KUBSan (CONFIG_UBSAN_TRAP=y) |
| CFI | KCFI (-fsanitize=kcfi, Clang 18) |
| Binary-only | QASAN (QEMU+ASan), DynamoRIO |
Property oracle patterns:
f(x) == f(f(x)){
"target": "linux/arm64",
"http": ":56700",
"workdir": "/path/to/workdir",
"kernel_obj": "/path/to/kernel",
"image": "/path/to/rootfs.ext3",
"sshkey": "/path/to/id_rsa",
"procs": 8,
"enable_syscalls": ["openat$module_name", "ioctl$IOCTL_CMD", "mmap"],
"type": "qemu",
"vm": { "count": 4, "cpu": 2, "mem": 2048 }
}
enable_syscalls to deepen coverage on specific subsystemssyz-extract to pull constants for custom modulesCONFIG_KASAN=y, CONFIG_KCFI=y, CONFIG_DEBUG_INFO_BTF=ykcov filters and syz_cover_filter to direct coverageTUN/TAP + pseudo-syscalls (syz_emit_ethernet)./scripts/decode_stacktrace.sh vmlinux ... < dmesg.logsyzkaller repro:
syz-execprog -repeat=0 -procs=1 -cover=0 -debug target.repro
WTF snapshot harness skeleton (mpengine.dll / mini-filter):
g_Backend->SetBreakpoint("nt!KeBugCheck2", [](Backend_t *Backend) {
const uint64_t BCode = Backend->GetArg(0);
Backend->Stop(Crash_t(fmt::format("crash-{:#x}", BCode)));
});
FilterConnectionPort fuzzing:
HANDLE hPort;
FilterConnectCommunicationPort(L"\\PortName", 0, NULL, 0, NULL, &hPort);
FilterSendMessage(hPort, fuzzData, sizeof(fuzzData), NULL, 0, &bytesReturned);
IOCTL fuzzing pattern:
HANDLE hDev = CreateFile(L"\\\\.\\DeviceName", GENERIC_READ|GENERIC_WRITE, ...);
DeviceIoControl(hDev, ioctlCode, inputBuf, inputLen, outBuf, outLen, &ret, NULL);
DRIVER_VERIFIER_DETECTED_VIOLATION (0xc4), IRQL_NOT_LESS_OR_EQUAL (0xa).symfix; !analyze -v; k; !heap -p -a @raxCross-platform mpengine.dll on Linux (loadlibrary + HF_ITER + Intel PT):
// Bypass Lua VM to avoid stability issues
insert_function_redirect((void*)luaV_execute_address, my_lua_exec, HOOK_REPLACE_FUNCTION);
for (;;) {
HF_ITER(&buf, &len);
ScanDescriptor.UserPtr = fmemopen(buf, len, "r");
__rsignal(&KernelHandle, RSIG_SCAN_STREAMBUFFER, &ScanParams, sizeof ScanParams);
}
# Full Rust fuzzing pipeline
cargo test # 1. property tests
cargo +nightly miri test # 2. UB via interpreter
cargo +nightly careful test # 3. runtime bounds checks
cargo fuzz run fuzz_target_1 -- -max_total_time=3600 # 4. libFuzzer crashes
RUSTFLAGS="--cfg loom" cargo test --release # 5. concurrency (if needed)
cargo fuzz coverage fuzz_target_1 # 6. coverage report
Focus unsafe blocks on: Vec::from_raw_parts, unchecked indexing, transmute size mismatches, pointer arithmetic, FFI integer truncation.
go test -fuzz=Fuzz -run=^$ ./...cargo-fuzz or honggfuzz-rs__builtin_return_address(0) for PC tracking)wasmtime-fuzz, wafl for differential fuzzing across V8/Wasmer/Wasmtime- name: Build with afl-clang-fast
run: CC=afl-clang-fast make -j
- name: Fuzz (smoke, 15 min)
run: timeout 15m afl-fuzz -i seeds -o findings -- ./target @@ || true
- name: Upload crashes
if: always()
uses: actions/upload-artifact@v4
with:
path: findings/**/crashes/*
Use ClusterFuzzLite for persistent continuous fuzzing; cache corpora between runs.
Linux:
ulimit -c unlimited && sysctl -w kernel.core_pattern=core.%e.%p
gdb -q ./target core.* -ex 'bt' -ex 'info reg' -ex q
addr2line -e ./target 0xDEADBEEF
Windows:
# Enable local dumps
New-Item 'HKLM:\SOFTWARE\Microsoft\Windows\Windows Error Reporting\LocalDumps' -Force
# PageHeap
gflags /p /enable target.exe /full
Kernel KASAN/KMSAN:
dmesg -T | egrep -i 'kasan|kmsan' -A 60
./scripts/decode_stacktrace.sh vmlinux /lib/modules/$(uname -r)/build < dmesg.log
Reproducibility: pin CPU governor, disable ASLR only where safe, fix RNG seeds, save input sequences in persistent mode, record binary hashes and sanitizer options with every crash.
| Tool | Use Case |
|---|---|
| AFL++ | General GreyBox, CmpLog, MOpt, Nyx |
| Honggfuzz | Intel PT, crash detection, HF_ITER |
| libFuzzer | In-process, source available |
| syzkaller | Linux/Windows kernel syscall fuzzing |
| wtf | Snapshot fuzzing, Windows targets |
| Nyx | AFL++ snapshot mode (Intel PT) |
| Snapchange | AWS snapshot fuzzing |
| LibAFL | Custom Rust fuzzing framework |
| AFLGo | Directed fuzzing to target BB/function |
| kAFL | Kernel + OS fuzzing |
| Jackalope | Binary coverage-guided (Windows/macOS) |
| cargo-fuzz | Rust libFuzzer integration |
| Atheris | Python fuzzing |
| Nautilus | Grammar-based fuzzing |
| AFLTriage | Automated crash triage |
| afl-cov | Coverage analysis for AFL++ |
| ClusterFuzz | Distributed fuzzing infrastructure |
评论 (0)
暂无评论,成为第一个评论者吧!