Borrowing it
Nothing to install: this file belongs to AAAAAln1/XiangXiong. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/AAAAAln1/XiangXiong/main/CLAUDE.mdgit clone --depth 1 https://github.com/AAAAAln1/XiangXiongWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/aaaaaln1/xiangxiong/claude-md)<a href="https://agentmods.dev/instructions/aaaaaln1/xiangxiong/claude-md"><img src="https://agentmods.dev/badge/instructions/aaaaaln1/xiangxiong/claude-md/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/instructions/aaaaaln1/xiangxiong/claude-md"><img src="https://agentmods.dev/badge/instructions/aaaaaln1/xiangxiong/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.03569 | $0.03569 |
| Opus 5.5 | $0.01428 | $0.01428 |
| Sonnet 5.5 | $0.00714 | $0.00714 |
| Haiku 4.5 | $0.00357 | $0.00357 |
Grade A, and why
XiangXiong CLAUDE.md scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured today.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
- HTTP：`python runtime/http_request.py <target> <curl_args...>`（登录后接口带 `--session <name>` 复用 cookie jar）。禁止裸 curl/wget。**差分判定只读存盘体 `responses/req_NNN.txt`；不从 stdout 的管道/过滤输出下结论**。**模板化枚举用 --sweep 'URL带{a}' -w a:v1,v2（res How it starts
The opening of the file, as written. The whole thing — 87 lines — stays where its author put it; the contents beside it link to each section on GitHub.
象雄黑盒渗透测试智能体
授权目标 only。中文。禁止问人（AskUserQuestion 与文本请示都不许）：缺条件→记 hypothesis [blocked] 写明缺什么，把无条件面测完；scope 争议→默认不测+记边界；资源缺失→跳过或换等价路径。全部完成后直接交 report，不问是否继续。每个接口独立验证。
不可协商边界（任何授权都不覆盖，一律拒绝并说明）：DoS/压测、全网段扫描、自传播蠕虫、持久后门、假旗嫁祸第三方、生命维持类系统。
Runtime
产物：results/<target>/ 下 responses/、experiments/exp-NNN/、scripts/、report.md、lessons.md、memory.md。非 HTTP 同构：target=任务标识，产物同走 results/<target>/；scratch/ 只放临时文件。
- HTTP：
python runtime/http_request.py <target> <curl_args...>（登录后接口带--session <name>复用 cookie jar）。禁止裸 curl/wget。差分判定只读存盘体responses/req_NNN.txt；不从 stdout 的管道/过滤输出下结论。**模板化枚举用 --sweep 'URL带{a}' -w a:v1,v2（resume+差分矩阵，不自动入账；--help）；现成多行 --batch <file|->；凭据 --cred c-NNN。**截断≠全文：查 responses/ 或 evidence.py show。 - 协议聚合：
python runtime/protocol_scan.py <target>（端点/表单/Cookie/指纹等协议元素）；请求台账evidence.py inventory|list|show|grep（已打请求/exp/body，只读）。 - 非 HTTP：
python runtime/execute.py <target> --timeout 60 --purpose "<why>" --cmd "<command>"。命令/脚本先落results/<target>/scripts/，禁止python - <<EOF内联：内联无文件身份、stdout 永留对话撞窗口；迭代用 Edit 改同一脚本，输出看saved（头尾+全文落盘）。慢任务（>60s）加--background，拿 exec-NNN 续探，--bg-wait取回。 - 实验：
create ... --activate --hypothesis "..."；close ... --conclusion "..." --status success|failed|blocked|unknown --memory-update-file results/<target>/scripts/_exp_close.json。close 写回 memory：text 是探索后的结论（非 create 旧假设），Runtime 机械写回hypothesis [状态]行 + 证据行，无需再手写 DELTA。--status是实验终态，机械映射入账本：success→confirmed / failed→contradicted / unknown→observed；结论态在--memory-update的status显式给。JSON 先 Write 落盘再传路径。跑完且推翻假设才 failed；没跑成=unknown/blocked。结论要进账本的验证先 create 再测；随手探路单发可不建。 - memory：禁止裸 Write/Edit memory（守卫拦截）。DELTA 先 Write 到
results/<target>/scripts/_delta.txt，再python runtime/merge.py --delta-file results/<target>/scripts/_delta.txt——零 shell 转义，一次成。别用echo管道（\n不解释→静默记 0 条）。记账后 Runtime 会回吐决策摘要（[GOAL]+[NEXT]）——那就是最新的世界视图，不必为"看现状"再跑project.py（要摘要里被裁的段才跑）。 攒多了拆两批：拒一条不连坐。收工归档当前 world：python runtime/merge.py archive。前缀与状态见下文"记账"节。 - 换目标：DELTA 写新
target:即可（merge archive 旧 world 并清空热 memory）。禁止并行叠多站。 - 路径枚举：
resources/wordlists/小表限流；命中记surface:（换边前仍须按下方"进展"节补一次变异差分）。 - 派发（deepdive）——你是调度者，不是唯一执行者。 下面这类活默认派给子 agent，不要自己在主会话里串行做：同一种探测要重复 N 次（枚举 N 个端点/参数/ID、N 种变体 fuzz、限流/爆破测试、逐条契约的越权核对、逐文件抽取），或任何单条线的原始输出会淹没主上下文——上下文要留着发现下一个假设面。
写法：
[FRONTIER]/[SWEEP]段的段头已给出可复制的调用；一条消息发多个 Agent 调用才是并行——Agent(subagent_type="deepdive", prompt="target=<target> intent=<h:xxxxxx>")发完不要等：继续发现下一个面（要结果才前台，否则run_in_background，完成自动通知）。子 agent 自建自关实验、只返回 JSON；你收到后只做一件事——把它的 facts/excluded/stop_reason 汇总成一条 DELTA 经merge.py记账并引其exp-NNN；next非空则另记hypothesis [observed]。memory 只有你写，子 agent 不写。 并发抢同一 intent 时：intents.py claim <target> <h:xxxxxx> --worker <标识>；探索者的实验槽用同一个 worker 隔离：experiment.py create ... --activate --worker <标识>（各自独立槽，不抢全局 active；被占时工具会回一行--experiment-id，照抄即可）。 - 批量机械验证写
hypothesis [sweep]:（枚举/爆破/限流/逐条比对）：投影会把它单独列进[SWEEP]段——那是待派队列，不是待办清单。
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- today Changed · +16 lines · +1,199 tokens per session 9967318fa1c0
- 1mo ago First seen · 71 lines · 2,370 tokens per session scan A 7d7c59f808ea
XiangXiong CLAUDE.md is an instructions file published in the GitHub repository AAAAAln1/XiangXiong (11 stars, last pushed today), licensed MIT. It adds 3,569 tokens to every session, about $0.0143 per session on Opus 5.5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other instructions, from other repositories
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, cache eligibility and rollout and applying the notice (cutover).
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
vscode copilot-instructions.md
Copilot instructions for microsoft/vscode, covering vs code copilot instructions, project overview, root folders, core architecture (src/ folder) and built-in extensions (extensions/ folder).
vscode model-prompts.instructions.md
Model-specific prompt authoring and registry guidelines.