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 instructions/astroicers/security-weekly-mcp/agents-mdgit clone --depth 1 https://github.com/astroicers/security-weekly-mcpWrote 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/astroicers/security-weekly-mcp/agents-md)<a href="https://agentmods.dev/instructions/astroicers/security-weekly-mcp/agents-md"><img src="https://agentmods.dev/badge/instructions/astroicers/security-weekly-mcp/agents-md.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 5.1 | $0.01902 | $0.01902 |
| Opus 5 | $0.00951 | $0.00951 |
| Sonnet 5 | $0.00380 | $0.00380 |
| Haiku 4.5 | $0.00190 | $0.00190 |
Grade A, and why
security-weekly-mcp AGENTS.md 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 5d 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.
How it starts
The opening of the file, as written. The whole thing — 88 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md — security-weekly-mcp 開發慣例(人機一體適用)
本檔為開發慣例的單一入口:任何 harness(Claude Code 或其他讀 AGENTS.md 的工具)與人類開發者讀到相同內容。專案總覽、MCP 工具清單、資料來源明細見 README.md;貢獻流程細節見 CONTRIBUTING.md。
目錄導覽
Monorepo(uv workspace),完整目錄樹見 README.md「專案結構」:
packages/glossary/— 資安術語庫,Git Submodule →astroicers/security-glossary-tw;terms/分類 YAML、meta/元資料(categories、style_guide)、pending/待審術語。packages/mcp-server/— MCP Server 主程式(security_weekly_mcp);工具模組在tools/,分glossary.py(術語庫)、news.py(新聞收集)、report.py(週報)。config/—sources.yaml(資料來源設定,含各來源啟停狀態)、search_templates.yaml(搜尋查詢模板)、writing_style.yaml(寫作風格)。output/reports/— 產出的週報(SEC-WEEKLY-YYYY-WW.json),經deploy-rss.yml發布為 HTML 至 GitHub Pages。scripts/— 自動化腳本(如generate_rss.py產 HTML/RSS)。skill/— Claude Code Skill 定義(自然語言介面的原始碼)。tests/— pytest 測試。.asp/— gate 渲染物(gate.sh、gitleaks.toml);單一事實源為根目錄asp-gate.yaml,渲染物勿手改。另含pr.md——asp-ng worker 出勤的 PR 描述快照(每票覆寫,main 恆為最近一票;asp-ng 決策 24/#183 已知邊界,人勿編輯)。
語言與風格
- 文件與 commit 訊息使用繁體中文;技術名詞保留英文。
- Python 遵循 PEP 8,以 ruff 檢查與格式化(
make lint/make format)。 - 週報與術語內容遵循
packages/glossary/meta/style_guide.yaml用詞規範(單一事實源),常見對照:駭客(非「黑客」)、惡意程式(非「病毒」)、特洛伊木馬程式(非「木馬」)、暴力破解(非「爆破」)、身份驗證用「驗證」(非「認證」)。
分支與提交
- 流:工作分支 → PR →
main(單線,無 develop;merge 前 CI 須綠)。 - 分支名以型別前綴,如
feature/*、fix/*、chore/*、ci/*。 - Commit 遵循 Conventional Commits:
feat|fix|docs|refactor|test|chore|report|ci(scope): 摘要(與asp-gate.yaml的 commit-format 檢查一致)。 report型別專用於每週週報產出 commit,如report: weekly security report 2026-W32。- 提交前跑本地 gate:
bash .asp/gate.sh(lint、gitleaks staged 秘密掃描、commit 格式);本地 gate 與 CI 跑同一份渲染物,判定一致由構造保證。
分支流(2026-08-21 起)
- develop = 整合層:asp-ng worker 的 PR 一律以 develop 為 base;CI 全套照跑。
- main = 部署面:push main 會觸發 deploy-rss(RSS 發布)——develop→main 的促升 PR 必由人審併,這是本 repo 的部署閘,不在任何自動化放權路徑上。
票與 PR
- PR 清楚描述變更內容、關聯對應 issue(
Closes #n)、確保 CI 綠後才請 review。 - 一 PR 一主題;超出範圍的想法開新 issue 而非擴大 diff。
- 術語庫內容變更(
terms/、meta/)請至security-glossary-tw提 PR;本 repo 的 PR 只更新 submodule 指標。
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.
- 5d ago First seen · 88 lines · 1,902 tokens per session scan A 71ca93d24560
security-weekly-mcp AGENTS.md is an instructions file published in the GitHub repository astroicers/security-weekly-mcp (0 stars, last pushed 2d ago), licensed MIT. It adds 1,902 tokens to every session, about $0.0095 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other instructions, from other repositories
glean CLAUDE.md
Claude Code instructions for LeslieLeung/glean, covering claude.md, project overview, quick start, start infrastructure (postgresql + redis + milvus) and start all services (api + worker + web).
glean AGENTS.md
AGENTS.md instructions for LeslieLeung/glean, covering repository guidelines, project structure & module organization, build, test, and development commands, coding style & naming conventions and testing guidelines.
mcp-threatintel AGENTS.md
Instructions for aplaceforallmystuff/mcp-threatintel, covering agents.md - mcp-threatintel, tech stack, architecture, development commands and environment variables.
mcp-panther CLAUDE.md
Instructions for panther-labs/mcp-panther, covering code guidelines, development documentation, validation guidelines, what fastmcp/pydantic handles automatically and when to use beforevalidator.
pentest-mcp-server AGENTS.md
AGENTS.md instructions for cyanheads/pentest-mcp-server, covering developer protocol, core rules, patterns, tool (representative — pentestlookuptechnique) and server instructions.
pentest-mcp-server CLAUDE.md
Claude Code instructions for cyanheads/pentest-mcp-server, covering developer protocol, core rules, patterns, tool (representative — pentestlookuptechnique) and server instructions.