SML Risk Firewall
Pre-trade risk checks for agent trades: orders vs your limits, signed receipts.
제작자가 쓴 설명(영어 원문)
Wraps the truesignal CLI as a single generic MCP tool for threat/connector status checks.
이 세 가지는 절대 하나의 점수로 합치지 않아요. 의견, 측정, 인기는 서로 다른 것이니까요.
이 서버는 아직 쉬운 말로 된 요약을 쓰지 못했어요. 아래는 레지스트리에서 읽어 온 사실이에요.
쓰고 있는 AI 앱을 고른 다음 설정을 복사하세요.
필요한 것
어떤 앱을 쓰세요?
위에서 앱을 고르세요. 기억해 두었다가 다음부터는 버튼에 “…에 설치”라고 표시해요.
Claude Desktop → Settings → Developer → Edit Config를 열고 claude_desktop_config.json에 이 내용을 붙여 넣으세요
{
"mcpServers": {
"truesignal": {
"command": "uvx",
"args": [
"truesignal-cli",
"truesignal-mcp"
]
}
}
}터미널에서 이 명령어를 실행하세요
claude mcp add truesignal -- uvx truesignal-cli truesignal-mcp프로젝트의 .cursor/mcp.json 또는 ~/.cursor/mcp.json에 붙여 넣으세요
{
"mcpServers": {
"truesignal": {
"command": "uvx",
"args": [
"truesignal-cli",
"truesignal-mcp"
]
}
}
}브라우저가 앱을 열지 물어본 다음, 앱이 설정을 보여 주고 확인을 요청해요. 키 값은 비워 두니 직접 입력하세요.
프로젝트의 .vscode/mcp.json에 붙여 넣으세요
{
"servers": {
"truesignal": {
"type": "stdio",
"command": "uvx",
"args": [
"truesignal-cli",
"truesignal-mcp"
]
}
}
}브라우저가 앱을 열지 물어본 다음, 앱이 설정을 보여 주고 확인을 요청해요. 키 값은 비워 두니 직접 입력하세요.
이 서버는 내 컴퓨터에서 실행되는데, ChatGPT는 인터넷에 있는 서버에만 연결할 수 있어요. 다른 앱을 고르세요.
선택한 앱은 이 기기의 쿠키에만 저장돼요. 계정은 필요 없어요.
이 설정은 제작자의 레지스트리 정보로 만들었어요. README와 다르면 README를 따르세요.
내 컴퓨터에서 실행돼요. 설치하면 AI 앱에 도구 목록이 나타나요.
누구나 볼 수 있는 정보를 점검했고, 코드는 읽지 않았어요. 해당하지 않는 항목은 개수에서 뺐어요. · 2026년 11월 7일 점검
레지스트리 이름이 GitHub 계정 RudrenduPaul 기준으로 확인되었고, 이 계정이 저장소를 소유하고 있어요
아직 GitHub에서 이 저장소를 읽어 오지 않았어요
아직 GitHub에서 이 저장소를 읽어 오지 않았어요
내 컴퓨터에서 실행되는 서버라 인터넷 엔드포인트가 없어요
내 컴퓨터에서 실행되며, 대신 로컬 설정을 사용해요
내 컴퓨터에서 실행돼요. 도구 목록은 인터넷 엔드포인트가 있는 서버에서만 읽을 수 있어요
레지스트리에서 이 카테고리의 활성 서버 전체와 비교했어요. 스타는 게시자가 저장소를 소유한 경우에만 세요.
해당하는 기본 점검 1개 중 1개를 통과했어요. 코드는 읽지 않았어요. 키나 토큰을 넣기 전에 서버가 어떤 권한을 요구하는지 확인하고, 꼭 필요한 권한만 있는 키를 쓰세요.
레지스트리에 필수 키가 나와 있지 않아요. 다만 처음 쓸 때 로그인을 요구하는 서버도 있어요.
내 컴퓨터에서 실행돼요. AI 앱이 필요할 때 실행하고, 데이터는 제작자의 서버를 거치지 않아요(서버 자체가 온라인 서비스를 호출하는 경우는 제외).
GitHub 저장소는 마지막으로 – 변경됐고, 최신 버전(0.1.5)은 3개월 전 레지스트리에 올라왔어요.
Claude Desktop, Claude Code, Cursor, VS Code처럼 MCP를 지원하는 앱에서 쓸 수 있어요. 내 컴퓨터에 Python (uv)도 있어야 해요.
같은 카테고리에는: SML Risk Firewall, Nife, ipdata MCP by usefulapi
같은 카테고리에서 다른 제작자가 만든 서버예요. 스타 순으로 정렬했어요.
카드 보는 법:★ GitHub 스타i세 기둥: 리뷰 · 점검 · 사용 정도i업데이트 = 최신 버전을 레지스트리에 공개한 날i
Pre-trade risk checks for agent trades: orders vs your limits, signed receipts.
Manage Nife.io cloud apps, sites, secrets and Git integrations from MCP clients.
Look up location, network, company and threat data for IP addresses and ASNs in ipdata.
제작자의 저장소에서 쓰인 그대로 가져왔어요. PopNx MCP가 검토하거나 보증한 내용이 아니에요. HTML, 이미지, 스크립트는 제거했어요. 링크를 누르면 이 사이트를 벗어나요.
Install • Quickstart • Features • CLI Reference • MCP Server • Compare • FAQ
A personal OSINT/security intelligence feed with a no-fabrication guarantee verified by 22 automated tests: every connector's failure path returns real cached data or nothing, never an invented data point.
TrueSignal ships as two independent, equally first-class packages -- pick whichever fits your toolchain, or install both:
npm (JS/TS CLI):
npx truesignal-cli init
truesignal-cli is published on npm (see the badge above). To build from source instead:
git clone https://github.com/RudrenduPaul/truesignal.git && cd truesignal && npm install && npm run build && node dist/cli.js init
Requires Node.js 18.17 or later. Both install paths verified working on 2026-08-03.
For repeat use, npm link after building gives you the truesignal command directly instead of typing node dist/cli.js.
pip (Python library + CLI):
pip install truesignal-cli
A genuine, independent Python port -- not a wrapper around the Node binary -- with the same five connectors, the same provenance-stamping guarantee, and the same init/feed/verify CLI surface. See python/README.md for the Python-specific quickstart. Both packages are maintained together; neither is deprecated in favor of the other.
$ truesignal init
truesignal connector status:
[ready] CISA Known Exploited Vulnerabilities (cisa-kev) -- no configuration needed [not configured] Cloudflare Radar (cloudflare-radar) -- set CLOUDFLARE_RADAR_API_TOKEN [not configured] Reddit (reddit) -- set REDDIT_CLIENT_ID, REDDIT_CLIENT_SECRET [not configured] Telegram (telegram) -- set TELEGRAM_BOT_TOKEN [ready] GDELT (gdelt) -- no configuration needed
2/5 connectors ready. Set the missing environment variables above to enable the rest. See .env.example. Next: run "truesignal feed" to see your feed now.
$ truesignal feed --source cisa-kev
[live] cisa-kev: CVE-2023-4346: KNX Association KNX Protocol Connection Authorization Option 1 Overly Restrictive Account Lockout Mechanism Vulnerability -- https://nvd.nist.gov/vuln/detail/CVE-2023-4346 -- 1d ago [live] cisa-kev: CVE-2026-46817: Oracle E-Business Suite Improper Privilege Management Vulnerability -- https://nvd.nist.gov/vuln/detail/CVE-2026-46817 -- 1d ago [live] cisa-kev: CVE-2026-15410: SonicWall SMA1000 Appliances Code Injection Vulnerability -- https://nvd.nist.gov/vuln/detail/CVE-2026-15410 -- 2d ago
This is a real, unedited capture (npm run build && node dist/cli.js init && node dist/cli.js feed --source cisa-kev), against the live CISA-KEV catalog, on 2026-07-15. CVE ids, urls, and ages are real.
Connector interface (src/truesignal/types.ts): CISA Known Exploited Vulnerabilities, Cloudflare Radar, Reddit (official OAuth API), Telegram (official Bot API), and GDELT.new Date()), and an explicit live or fallback status. A fallback item also carries fallbackAgeSeconds, so you always know exactly how stale what you're looking at is.Math.random(), a fake-data library, or a "rewrite the timestamp to now" path. A dedicated test suite (src/truesignal/provenance/no-fabrication.test.ts, 22 tests) proves this for every connector's live, fallback, and empty-cache-failure paths, and a static check scans every connector source file for forbidden patterns.--json flag with a stable, documented shape and real, distinct exit codes, so an agent can parse a result without scraping human-formatted text.truesignal init and truesignal feed work the moment the CLI is installed..env files or phone home.Connector and one line in connectors/index.ts -- never a change to the CLI or provenance layer. See CONTRIBUTING.md.truesignal init
truesignal feed
init tells you which connectors are ready right now (CISA-KEV and GDELT need nothing) and which environment variables are still missing for the rest. feed pulls from every configured connector. Real, unedited output from the two zero-config sources:
[live] cisa-kev: CVE-2023-4346: KNX Association KNX Protocol Connection Authorization Option 1 Overly Restrictive Account Lockout Mechanism Vulnerability -- https://nvd.nist.gov/vuln/detail/CVE-2023-4346 -- 1d ago
[fallback, 37m old] gdelt: White House launches AI-backed Gold Eagle initiative for cybersecurity vulnerability coordination -- https://www.scworld.com/brief/white-house-launches-ai-backed-gold-eagle-initiative-for-cybersecurity-vulnerability-coordination -- 2h ago
That second line is real, too: on this run GDELT's live fetch didn't complete, so truesignal showed the last real GDELT items it had actually fetched, honestly labeled fallback with their real age, instead of being silently repainted as current.
To enable the other three sources (Cloudflare Radar, Reddit, Telegram), copy .env.example to .env, fill in real free credentials, then export them into your shell -- truesignal doesn't auto-load .env files:
set -a && source .env && set +a
Generated directly from this build's actual --help output (node dist/cli.js --help and each subcommand's --help).
truesignalUsage: truesignal [options] [command]
A provenance-first OSINT/security intelligence feed. Every item carries a real source URL, a real timestamp, and an explicit live/fallback flag -- never a fabricated or silently-replayed data point.
Options: -V, --version output the version number -h, --help display help for command
Commands: init [options] Check which connectors are ready to use right now and which environment variables are still needed for the rest. CISA-KEV and GDELT need no configuration -- truesignal works with zero setup for those two sources. feed [options] Pull the current feed from every configured connector, or one connector with --source. Prints human-readable output by default; use --json for a stable, agent-parseable schema. verify [options] Re-fetch the source connector named in and confirm whether that item still resolves to real, live provenance, has fallen back to cached data, or can no longer be found. help [command] display help for command
truesignal init [--json]Exit code 0 if at least one connector is usable, 2 if none are (shouldn't happen -- CISA-KEV and GDELT need no configuration).
truesignal feed [--source ] [--json]--source <name> only pull from this connector, e.g. cisa-kev, gdelt
--json print machine-readable JSON instead of human-readable text
Exit codes: 0 success, 2 no connectors configured to run, 3 every configured connector's fetch failed with no data to show.
--json output is a stable, documented FeedItem[] shape. Real capture:
{
"items": [ { "id": "cisa-kev:CVE-2023-4346", "source": "cisa-kev", "title": "CVE-2023-4346: KNX Association KNX Protocol Connection Authorization Option 1 Overly Restrictive Account Lockout Mechanism Vulnerability", "url": "https://nvd.nist.gov/vuln/detail/CVE-2023-4346", "timestamp": "2026-07-15T00:00:00.000Z", "status": "live" } ] }
truesignal verify [--json]![truesignal verify cisa-kev:CVE-2023-4346 confirming live provenance, followed by truesignal feed --source cisa-kev --json printing the stable FeedItem[] JSON schema, run via npx truesignal-cli](./docs/usage.gif)
Re-fetches the source named in (format : , e.g. cisa-kev:CVE-2023-4346) and confirms whether that item still resolves to real, live provenance, has fallen back to cached data, or can no longer be found. Real capture:
$ truesignal verify cisa-kev:CVE-2023-4346
cisa-kev:CVE-2023-4346: LIVE -- https://nvd.nist.gov/vuln/detail/CVE-2023-4346 -- 2026-07-15T00:00:00.000Z
Exit codes: 0 found and live/fallback, 1 re-fetched successfully but the item is gone, 2 the connector isn't configured, 3 the re-fetch failed, 4 the item id is malformed or names an unknown source.
TrueSignal ships a Model Context Protocol (MCP) server, so an MCP-compatible agent runtime (Claude Desktop, Claude Code, or any other MCP client) can call TrueSignal directly instead of shelling out to the CLI and parsing text.
pip install "truesignal-cli[mcp]"
Add it to your Claude Desktop config (claude_desktop_config.json):
{
"mcpServers": { "truesignal": { "command": "truesignal-mcp" } } }
The server exposes a single tool, run(args: list[str]), that shells out to the installed truesignal CLI with the given arguments and returns its parsed JSON output -- for example, run(args=["feed", "--source", "cisa-kev", "--json"]) returns the parsed FeedItem[] JSON for the CISA-KEV connector's current feed, the same data truesignal feed --source cisa-kev --json prints on the command line. Every failure mode (missing CLI, timeout, non-zero exit, unparseable output) is caught and returned as {"error": ...} instead of raising. See python/src/truesignal/mcp_server.py for the implementation.
Every cell below is a cited, checkable fact, current as of 2026-08-03, drawn only from tools actually run in this pass. See benchmarks/ for the reproduction command behind every TrueSignal number.
| | TrueSignal | Crucix
앞부분만 보여 드려요. 나머지는 저장소에서 읽어 보세요