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 agentmods add skills/unclecheng-li/deepsec/secknowledge-skillnpx skills add Unclecheng-li/DeepSec --skill secknowledge-skillgit clone --depth 1 https://github.com/Unclecheng-li/DeepSecWrote 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/unclecheng-li/deepsec/secknowledge-skill)<a href="https://agentmods.dev/skills/unclecheng-li/deepsec/secknowledge-skill"><img src="https://agentmods.dev/badge/skills/unclecheng-li/deepsec/secknowledge-skill.svg" alt="Measured on agentmods" 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 | $0.00321 | $0.04437 |
| Opus 5 | $0.00161 | $0.02218 |
| Sonnet 5 | $0.00064 | $0.00887 |
| Haiku 4.5 | $0.00032 | $0.00444 |
Grade A, and why
secknowledge-skill scanned grade A with 0 findings 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 4d 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.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
This is a copy
100% identical to secknowledge-skill — 0 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 241 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Web 和 AI 安全测试知识库
知识源: WooYun 88,636 漏洞 × 先知 5,600+ 文档 × GAARM 150 AI 风险 × OWASP 架构: SKILL.md（路由）→ references/（按场景加载）
VulnClaw 集成说明
- 本 Skill 集成自
Pa55w0rd/secknowledge-skill，在 VulnClaw 中作为secknowledge-skillspecialized skill 使用；上游声明为 MIT License。 - CTF/SRC 场景先加载
references/vulnclaw-ctf-src-routing.md判断资料入口，再按漏洞类型加载web-*、ai-*、testing-methodology.md或gaarm-risk-matrix.md。 - 与 VulnClaw 现有技能协同：CTF 单题技巧优先结合
ctf-web/ctf-crypto/ctf-misc，SRC 和实战漏洞挖掘优先使用本 Skill 的方法论、案例映射、风险矩阵和证据约束。 - 输出时保留上游的授权边界、引用标注和"假设/确认"区分；未能从 reference 佐证的 payload、CVE、GAARM/OWASP 编号必须明确标注为未核对。
触发条件
触发条件（AND 组合）：
- 用户意图是执行安全测试（渗透/挖洞/利用/审计） — 非讨论/学习
- 提供了具体目标：URL、接口、代码片段、模型/Agent 架构、MCP 配置 — 非抽象问题
- 任务涉及以下领域之一：
- Web: SQL 注入/XSS/命令执行/越权/文件上传/SSRF/反序列化/XXE/GraphQL/HTTP 走私
- AI: Prompt 注入/越狱/MCP 投毒/Agent 滥用/RAG 投毒/沙箱逃逸/模型窃取
- 绕过: WAF/内容过滤/Guard Rails 绕过
不触发（任一命中即路由到他处）：
- 概念讲解："什么是…"、"…原理"、"…怎么防御" → 普通问答
- 非安全代码审查："review 代码质量"、"优化性能" → 普通 code review
- 业务 bug：语法错误、空指针、业务逻辑错误（非安全逻辑）→ 普通 debug
- 深度白盒代码审计（Source-Sink 污点传播、AST 分析）→ code-audit-skill
- 查 CVE 文档、工具文档 → WebSearch/Context7
歧义处理：目标和意图不明时，先问："目标是什么？你希望做渗透测试 / 代码审计 / 还是了解概念？"
行为准则（整个会话有效，不因对话长度放松）
- ❗ 所有 Payload/CVE 编号/风险编号必须引用 reference 文件的具体章节 — 每次输出前自检。未在 reference 中的一律标注 "UNABLE TO CITE"，禁止编造。
- ❗ 区分"漏洞假设"与"漏洞确认" — 基于方法论推断的潜在风险 → 标注
假设（需验证）；有明确证据的 → 标注已确认（证据: …）。禁止混淆。 - ❗ 授权边界 — 任何利用步骤输出前必须确认是 CTF/授权渗透/本人环境。无授权上下文只输出分析，不输出可直接武器化的完整 Payload。
幻觉防护与来源引用
| 内容类型 | 正确输出 | 禁止输出 |
|---|---|---|
| CVE 编号 | 引用具体 reference 文件和章节，或标 "UNABLE TO CITE — 建议 WebSearch 核实" | 编造 CVE-YYYY-NNNN |
| Payload | 从 references/web-*.md 或 references/ai-*.md 内 payload 章节引用 |
凭印象写 payload |
| GAARM 风险编号 | 引用 references/gaarm-risk-matrix.md |
自造编号 |
| OWASP 条目 | LLM01-10 / ASI01-10 / WSTG-* 引用 testing-methodology.md §10.x |
改写编号含义 |
| 工具/命令 | 仅使用在 reference 中出现过的，或明确标注 "通用命令（未在 reference 中核对）" | 伪造工具参数 |
| 无检索结果 | "UNABLE TO ASSESS：reference 未覆盖此场景，建议 WebSearch" | 凭经验推测作为结论 |
What ships with it
40 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- references/ai-app-agent-cot.md 19 KB
- references/ai-app-deploy.md 4.5 KB
- references/ai-app-frontier.md 6.2 KB
- references/ai-app-mcp.md 8.6 KB
- references/ai-app-prompt.md 19 KB
- references/ai-app-train.md 15 KB
- references/ai-baseline-app.md 9.6 KB
- references/ai-baseline-deploy.md 19 KB
- references/ai-baseline-escape.md 6.5 KB
- references/ai-baseline-train.md 6.4 KB
- references/ai-data-app.md 33 KB
- references/ai-data-deploy.md 7.1 KB
- references/ai-data-train.md 22 KB
- references/ai-identity-app.md 33 KB
- references/ai-identity-deploy.md 7.1 KB
- references/ai-identity-train.md 5.1 KB
- references/ai-model-content.md 23 KB
- references/ai-model-copyright.md 4.8 KB
- references/ai-model-deploy.md 5.7 KB
- references/ai-model-extraction.md 12 KB
- references/ai-model-hallucination.md 9.6 KB
- references/ai-model-jailbreak.md 13 KB
- references/ai-model-misuse.md 19 KB
- references/ai-model-train.md 10 KB
- references/gaarm-risk-matrix.md 14 KB
- references/testing-methodology.md 25 KB
- references/vulnclaw-ctf-src-routing.md 3.9 KB
- references/web-deployment-security.md 15 KB
- references/web-deser.md 4.3 KB
- references/web-leak.md 3.3 KB
- references/web-logic-auth.md 16 KB
- references/web-modern-protocols.md 10 KB
- references/web-rce.md 6.9 KB
- references/web-sqli-fushuling-one-pass.md 7.7 KB
- references/web-sqli.md 6.9 KB
- references/web-ssrf-misc.md 6.0 KB
- references/web-traversal.md 3.5 KB
- references/web-upload.md 5.3 KB
- references/web-xss.md 5.0 KB
- references/web-xxe.md 3.1 KB
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.
- 4d ago First seen · 241 lines · 321 tokens per session scan A 7b6b2e78861f
secknowledge-skill is a skill published in the GitHub repository Unclecheng-li/DeepSec (386 stars, last pushed 11d ago), licensed MIT. It adds 321 tokens to every session and 4,437 once invoked, about $0.0016 per session on Opus 5. A static security scan graded it A with 0 findings. It is 100% identical to secknowledge-skill, differing in 0 lines, and is treated as a copy.
Other skills, from other repositories
defense-evasion
Endpoint defense bypass — AMSI/ETW patching, ScareCrow framework, custom loaders, direct/indirect syscalls, LOLBAS execution, process injection.
opsec
Operational security management — traffic shaping, scan rate limiting, source IP management, tool signature avoidance, evidence handling, anti-detection patterns.
benchmark
Benchmark mode marker — engagement objective is flag capture. Generic engagement rules apply unchanged.
seven-question-gate
7-question gate run before promoting a finding to FINDING + opening a report. Kills weak/non-impactful findings before they reach the report stage and damage validity ratio.
finding-protocol
Operational-tier finding template — minimal fields for sub-agent decision support. Heavyweight deliverable promotion lives in skills/decepticon/final-report.
bloodhound-bhce
Operate BloodHound Community Edition v9.2.2 via Decepticon's bhce tools — health check, Cypher passthrough, SharpHound ZIP ingest. Replaces the in-house ingest + ESC post-process pipeline per ADR-0005.