agentmods.dev

dym-cti-expert-safety-gates

dym-cti-expert-safety-gates is a skill for Claude Code from Rheinmir/setup. It costs 61 tokens per session (1,853 once invoked), scanned A, original, MIT.

A set of safety hooks that control risky actions in a coding-agent session. They can block sensitive case data from being written, ask before irreversible external actions, and report session details.

In plain words
What is it for?
Use it to protect tracked files from investigation data, require approval for outbound commands or tools, and monitor the session's configured safeguards.
Why use it?
It makes safety checks explicit so you know what is blocked, what requires confirmation, and what is only reported.

Skill for Claude Code

Written for Claude Code: disable-model-invocation in frontmatter. Also seen: mentions CLAUDE.md; names the NotebookEdit tool; mentions Claude Code.

Needs its repository: it runs a file that does not travel with it, so clone the repository first. The line is python3 tests/test_hooks.py.

not rated 11repo 20d ago A scan 61 tokens original MIT

Good fit Use it to protect tracked files from investigation data, require approval for outbound commands or tools, and monitor the session's configured safeguards.

Compare 6 skills from other repositories ↓
View source ↗ Rheinmir/setup
Install

Getting it into your agent

It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.

Clone the repo
git clone --depth 1 https://github.com/Rheinmir/setup
agentmods
npx agentmods add skills/rheinmir/setup/dym-cti-expert-safety-gates

Made for: Claude Code.

Wrote 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.

agentmods badge for dym-cti-expert-safety-gates

README.md
[![agentmods](https://agentmods.dev/badge/skills/rheinmir/setup/dym-cti-expert-safety-gates/github.svg)](https://agentmods.dev/skills/rheinmir/setup/dym-cti-expert-safety-gates)
Your own site
<a href="https://agentmods.dev/skills/rheinmir/setup/dym-cti-expert-safety-gates"><img src="https://agentmods.dev/badge/skills/rheinmir/setup/dym-cti-expert-safety-gates/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.

agentmods 80×15 button for dym-cti-expert-safety-gates

Your own site · 80×15
<a href="https://agentmods.dev/skills/rheinmir/setup/dym-cti-expert-safety-gates"><img src="https://agentmods.dev/badge/skills/rheinmir/setup/dym-cti-expert-safety-gates.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 61 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,853 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00061 $0.01853
Opus 5.5 $0.00024 $0.00741
Sonnet 5.5 $0.00012 $0.00371
Haiku 4.5 $0.00006 $0.00185

Measured yesterday against content hash 55eff483648f, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

dym-cti-expert-safety-gates 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 yesterday.

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.

.overstack/doyourmagic/cti-expert/skills/dym-cti-expert-safety-gates/SKILL.md · 111 lines

How it starts

The opening of the file, as written. The whole thing — 111 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Skill: dym-cti-expert-safety-gates — Rào an toàn: cái gì bị CHẶN, cái gì HỎI, cái gì im lặng

Vì sao dùng: để bạn không ngạc nhiên giữa case, và quan trọng hơn — để bạn không tắt nhầm một rào. Sinh ra cái gì: không sinh gì; nó đổi cách bạn đọc mọi prompt xác nhận mà cti-expert bật lên.

Ba hook, và chúng chỉ tồn tại khi cài theo đường plugin

hooks/hooks.json đăng ký đúng ba mục:

Hook Bắt sự kiện Hành vi
leakguard.py PreToolUse · Write|Edit|NotebookEdit deny — chặn ghi dữ liệu case vào file tracked
actionguard.py PreToolUse · Bash|mcp__.* ask — hỏi trước hành động outbound không thu hồi được
sessionguard.py SessionStart báo tier backend + cảnh báo khi số @tool đổi

bash scripts/register.sh không cài được hook. Chỉ /plugin install cti-expert mới có (01 bước 3b). Verify bằng /hooks trong Claude Code — phải thấy 3 mục.

leakguard.py — RULE 1 tại thời điểm GHI

Repo cấm dữ liệu điều tra lọt vào file tracked: tên người thật, hạ tầng của case, GA4/GTM thật, case ID, PII của operator — kể cả dưới dạng "ví dụ minh hoạ" trong comment hay fixture test.

Có hai lớp thi hành, cố ý:

  1. scripts/leakcheck.sh chạy ở git pre-commit (cài bằng bash scripts/install-hooks.sh, phải chạy mỗi lần clone mới vì .git/hooks/ không được track).
  2. hooks/leakguard.py chạy ở thời điểm Write/Edit.

Lý do có cái thứ hai: hook git chỉ nổ lúc commit và git commit --no-verify bỏ qua được, trong khi một agent thì ghi liên tục và commit hiếm khi. Ở tầng Write không có cờ --no-verify nào cả.

leakguard.py không tự cài lại các mẫu regex — nó gọi ra scripts/leakcheck.sh. Một bản sao thứ hai sẽ trôi lệch, và một guard đã trôi lệch thì luôn báo sạch.

Nó chỉ deny khi cả hai đúng: đường dẫn nằm trong một checkout cti-expert, và git không ignore đường dẫn đó. Ghi vào intel_engine/cases/, intel_engine/knowledge/, .env — tức các kho đã gitignore — hoàn toàn không bị đụng. Ghi ngoài checkout cũng vậy.

Read the full file on GitHub · 111 lines

Changes

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.

  1. yesterday First seen · 111 lines · 61 tokens per session scan A 55eff483648f

Subscribe to this mod's changes

dym-cti-expert-safety-gates is a skill published in the GitHub repository Rheinmir/setup (11 stars, last pushed 20d ago), licensed MIT. It adds 61 tokens to every session and 1,853 once invoked, about $0.0002 per session on Opus 5.5. 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 2026-11-07.

Building agentsSecurity
Related

dym-cti-expert-safety-gates compared with skills from other repositories

dym-cti-expert-safety-gates next to 6 skills from other repositories
ModTypeStarsLast pushLicenceScanTokensWorks with
dym-cti-expert-safety-gatesRheinmir/setupSkill1120d agoMITA61Claude Code
imagegenopenai/codexSkill134ktodayApache-2.0A113Codex
local-ai-agentsmicrosoft/ai-agents-for-beginnersSkill79kyesterdayMITA200Claude Code, Codex
playwright-climicrosoft/playwrightSkill100kyesterdayApache-2.0A19Claude Code
browser-automationopenclaw/openclawSkill394ktodayMITA36Claude Code
systematic-debuggingobra/superpowersSkill314k4d agoMITA21Claude Code
next-cache-components-adoptionvercel/next.jsSkill143ktodayMITA95Claude Code