Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add SeaOf0/dsh-redteam-model --skill ctf-playbookgit clone --depth 1 https://github.com/SeaOf0/dsh-redteam-modelWrote 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/skills/seaof0/dsh-redteam-model/ctf-playbook)<a href="https://agentmods.dev/skills/seaof0/dsh-redteam-model/ctf-playbook"><img src="https://agentmods.dev/badge/skills/seaof0/dsh-redteam-model/ctf-playbook/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/skills/seaof0/dsh-redteam-model/ctf-playbook"><img src="https://agentmods.dev/badge/skills/seaof0/dsh-redteam-model/ctf-playbook.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00367 | $0.05443 |
| Opus 5 | $0.00183 | $0.02721 |
| Sonnet 5 | $0.00073 | $0.01089 |
| Haiku 4.5 | $0.00037 | $0.00544 |
Grade A, and why
ctf-playbook 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 7d ago.
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.
- 通用：python3（pwntools/z3/pycryptodome/requests）、file/strings/objdump、curl/wget、git How it starts
The opening of the file, as written. The whole thing — 200 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CTF 解题作战手册
主观念=flag 真实性主线；两门 board/flag；成果页=ledger 台账板式（复用）。
开工顺序：工作区发现 → WORKSPACE.md → tool-plane 检测登记 → 优先看 route-boost 信封 （已含门禁与 canonical 名），信封缺失/不确定再调 gates_list（mode=ctf-solver）。
覆盖度台账（operation-state 扩展，与门禁同源）：
operation_goal登记目标契约后先operation_constraints登记用户约束（deny/allow 每行一条，带匹配词的 deny 命中 bash/fetch 即确定性拦；约束每轮进信封防压缩丢失）再operation_scope登记范围分母——每行一项（资产/路由/模块/账号/题目等目标实际要求覆盖的单元；「id: 标签」可固定 id；最小范围原则：只登记目标明确点到或派生必需的面，绝不擅自放大）；每测完一项即operation_progress tested=<id> evidence=<evidence 编号/矩阵行/输出文件>记分子（幂等，重复标记刷新证据）。scope 登记后本模式报告门自动开启算术对账：报告须含「覆盖：M/N」声明行且与台账实测一致——部分覆盖照实声明可过，虚报/漏报拦门。开新方向（派单/追线/阶段切换）先operation_intent登记带锚（anchor=boot 开局豁免 / criterion 准则 / scope 范围 / finding 本会话成果 / chain 链路节点 + id）——方向只能锚在已确立的证据上；收口operation_progress intent_done/intent_blocked/intent_dropped（blocked/dropped 附原因）；未收口意图拦报告落盘。
定位与设计依据
CTF 解题模式（ctf-solver）是轻量解题台：题目与题目环境默认沙盒内解题（题目环境=授权解题
对象），flag 是唯一可交付物。知识层=预设内建 refs/ 知识库（118 篇，模块化：web/pwn/
reverse/crypto/forensics/misc/ai-ml/osint/malware + solve-challenge 分诊入口），跨模块面
（AD/云/mobile）按生态协作就地加载对应专业模式 refs。与评估类模式的差异：无授权评估语义、
无六字段报告、无检测缺口——本模式只做「解题编排 + 赛制策略 + flag 台账 + 复盘」。
flag 真实性主线（主观念）
- flag 真实 = 竞赛平台提交回显通过 / 本地验证脚本（check 器）通过；不猜不撞不伪造。
- 题面是出题人与你的唯一契约：线索优先，每题先梳理题面再动手。
- 每题闭环：题面 → 假设 → 验证 → flag；未解题目如实登记（进展/卡点/已尝试路径）。
- 猜测性 flag 标「待验证」，绝不标记已解。
解题流程（四阶段 ↔ 两门）
| 阶段 | 产物（canonical） | 门 |
|---|---|---|
| 1 题面登记 | challenge-board.md（题名/模块/分值/线索梳理/状态，≥1 行表）+ evidence-index.md | board |
| 2 模块路由与解题 | 每题工作目录（exp/<题名>/：脚本与中间产物）+ 台账行更新 | —（解题循环内） |
| 3 flag 验证与台账 | flag-ledger.md（题名/模块/flag/验证证据/状态，≥1 行表） | flag |
| 4 复盘报告 | CTF 解题报告（$file） | flag |
阶段纪律：board 过门才开题；flag 过门才写报告。报告先落工作区根目录 → 过 flag 门 → 再复制进 reports/。所有 file 参数必须传绝对路径。
两门门禁
| 门 | 结构校验物（canonical） | 语义（manual，总控/复核员判定） |
|---|---|---|
| board 题面登记 | challenge-board.md（含标记：题名/模块/线索，≥1 行表，每行 ≥3 格）+ evidence-index.md（含字面标记 tool-plane、MCP，且 ≥1 行表） |
每行线索已梳理、模块判定合理 |
| flag 台账收口 | flag-ledger.md（含标记：flag/验证/状态，≥1 行表，每行 ≥4 格） | 每个「已解」flag 带验证证据；未解标卡点 |
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.
- 7d ago First seen · 200 lines · 367 tokens per session scan A b6ec624e4d64
ctf-playbook is a skill published in the GitHub repository SeaOf0/dsh-redteam-model (354 stars, last pushed yesterday), licensed MIT. It adds 367 tokens to every session and 5,443 once invoked, about $0.0018 per session on Opus 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 2026-09-03.
Other skills, from other repositories
infographic
Turn text, a document, or a topic into a graphic-first explainer: an imaginative schoolbook-style HTML page where full-page drawings, diagrams and legends carry the ideas and the text is titles only (a print-ready PDF only when asked). Use for an infographic, visual explainer, one-pager, data poster, concept or…
teach
A structured teaching workflow that builds lessons and keeps track of what a user has learned in the current workspace.
promentor
A tool that turns a software project into a hands-on course in the style of an MIT course. It can produce lessons, code walkthroughs, practical labs, behavior tests, grading, explanations, and AI reviews.
taptap-maker-dev-kit-guide
Explain the local TapTap Maker AI dev kit installed during Maker project clone. Use after a Maker project is cloned or bound, or when the user asks what CLAUDE.md, examples, templates, or urhox-libs are for.
quantum-error-correction
A clear and practical skill guide for learning and running a PennyLane-based qLDPC tutorial, from classical LDPC basics to CSS and Hypergraph Product code construction. Skill-first for covered code generation, runnable examples, execution, debugging, validation, and fixed workflows.
teach
A Chinese-language teaching skill for learning a new concept or ability through several lessons in the current workspace.