Fournit aux agents IA un accès à la threat intelligence, incluant plus de 1 700 menaces, 47 000+ IOC, des règles de détection et une cartographie MITRE ATT&CK.
Aussi connu sous threadlinqs-cmd/intelthreadlinqs-mcp
- Étoiles
- 0
- Outils
- 3
- Téléchargements, 30 jours
- —
- Dernière version
- 8.1.1
- Dernier commit
- 23 août 2026
- Licence
- MIT
Qu'est-ce que Threadlinqs Intelligence MCP ?
Fournit aux agents IA un accès à la threat intelligence, incluant plus de 1 700 menaces, 47 000+ IOC, des règles de détection et une cartographie MITRE ATT&CK. Il expose 3 outils, comme search_actors, get_actor, get_similar_threats.
Résumé généré à partir du README du projet
À propos du MCP server Threadlinqs Intelligence
Les analystes en sécurité et les développeurs peuvent doter les agents IA d'une connaissance approfondie de la threat intelligence en connectant ce serveur à des clients tels que Claude Desktop, Cursor ou Claude Code. Il offre un accès à un vaste référentiel contenant plus de 1 700 menaces, 47 000 indicateurs de compromission et 16 000 règles de détection formatées pour Splunk, Microsoft KQL et Sigma. Le serveur fonctionne via stdio et nécessite une variable d'environnement THREADLINQS_API_KEY pour fonctionner.
L'accès aux 81 outils disponibles est réservé aux utilisateurs disposant d'un abonnement Purple ou Gold. Bien que le serveur démarre et permette aux clients d'inspecter le catalogue d'outils avec des clés de niveaux inférieurs, la récupération effective des données est restreinte côté serveur. Les utilisateurs peuvent utiliser des outils spécialisés comme get_threat_hunting_bundle pour obtenir des dossiers complets ou get_mitre_gap_analysis pour identifier les lacunes de détection. Le système inclut également 25 prompts de type analyst-playbook pour aider les agents à orchestrer des workflows complexes, tels que le triage de CVE ou le profilage d'acteurs.
Extrait du README du projet. Chaque cas d'utilisation est vérifié par rapport à une citation du document.
Comment installer Threadlinqs Intelligence MCP
Formats vérifiés le 10 nov. 2026Ajouter à CursorAjouter à GooseInstaller Threadlinqs Intelligence MCP dans Claude Code
$ claude mcp add --env 'THREADLINQS_API_KEY=<YOUR_THREADLINQS_API_KEY>' intelthreadlinqs-mcp -- npx -y intelthreadlinqs-mcpInstaller Threadlinqs Intelligence MCP dans Cursor
~/.cursor/mcp.json
{
"mcpServers": {
"intelthreadlinqs-mcp": {
"command": "npx",
"args": [
"-y",
"intelthreadlinqs-mcp"
],
"env": {
"THREADLINQS_API_KEY": "<YOUR_THREADLINQS_API_KEY>"
}
}
}
}Installer Threadlinqs Intelligence MCP dans VS Code
.vscode/mcp.json
{
"servers": {
"intelthreadlinqs-mcp": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"intelthreadlinqs-mcp"
],
"env": {
"THREADLINQS_API_KEY": "<YOUR_THREADLINQS_API_KEY>"
}
}
}
}Installer Threadlinqs Intelligence MCP dans VS Code CLI
$ code --add-mcp '{"name":"intelthreadlinqs-mcp","type":"stdio","command":"npx","args":["-y","intelthreadlinqs-mcp"],"env":{"THREADLINQS_API_KEY":"<YOUR_THREADLINQS_API_KEY>"}}'Installer Threadlinqs Intelligence MCP dans Claude Desktop
claude_desktop_config.json
{
"mcpServers": {
"intelthreadlinqs-mcp": {
"command": "npx",
"args": [
"-y",
"intelthreadlinqs-mcp"
],
"env": {
"THREADLINQS_API_KEY": "<YOUR_THREADLINQS_API_KEY>"
}
}
}
}Installer Threadlinqs Intelligence MCP dans Codex
$ codex mcp add intelthreadlinqs-mcp --env 'THREADLINQS_API_KEY=<YOUR_THREADLINQS_API_KEY>' -- npx -y intelthreadlinqs-mcpInstaller Threadlinqs Intelligence MCP dans Codex config.toml
~/.codex/config.toml
[mcp_servers.intelthreadlinqs-mcp] command = "npx" args = ["-y", "intelthreadlinqs-mcp"] [mcp_servers.intelthreadlinqs-mcp.env] THREADLINQS_API_KEY = "<YOUR_THREADLINQS_API_KEY>"
Installer Threadlinqs Intelligence MCP dans Gemini CLI
$ gemini mcp add -s user -e 'THREADLINQS_API_KEY=<YOUR_THREADLINQS_API_KEY>' intelthreadlinqs-mcp npx -- -y intelthreadlinqs-mcpInstaller Threadlinqs Intelligence MCP dans Gemini settings.json
~/.gemini/settings.json
{
"mcpServers": {
"intelthreadlinqs-mcp": {
"command": "npx",
"args": [
"-y",
"intelthreadlinqs-mcp"
],
"env": {
"THREADLINQS_API_KEY": "<YOUR_THREADLINQS_API_KEY>"
}
}
}
}Outils MCP Threadlinqs Intelligence 3
Détectés depuis le READMEsearch_actorsget_actorget_similar_threats
| search_actors | — | |
| get_actor | — | |
| get_similar_threats | — |
Configuration
THREADLINQS_API_KEYRequisSecretThreadlinqs API key (Purple or Gold tier, tl_...) — get one at https://intel.threadlinqs.com/profile
THREADLINQS_API_URLOverride the API base URL (defaults to https://intel.threadlinqs.com)
| Nom | Description | Type |
|---|---|---|
| THREADLINQS_API_KEY | Threadlinqs API key (Purple or Gold tier, tl_...) — get one at https://intel.threadlinqs.com/profile | RequisSecret |
| THREADLINQS_API_URL | Override the API base URL (defaults to https://intel.threadlinqs.com) |
Vérifications automatisées
Exécuté le 14 nov. 2026Rapport complet →A passé nos vérifications automatisées. Il ne s'agit pas d'un audit de sécurité : lisez le code et donnez au serveur l'accès le plus restreint dont il a besoin. Fonctionnement des vérifications
- Dépôt source lié et accessiblethreadlinqs-cmd/intelthreadlinqs-mcp is public and reachablegithub.com/threadlinqs-cmd/intelthreadlinqs-mcp ↗Réussi
- Le nom du package correspond au dépôt sourceintelthreadlinqs-mcp names threadlinqs-cmd/intelthreadlinqs-mcp as sourcenpmjs.com/package/intelthreadlinqs-mcp ↗Réussi
- Provenance du build (attestation Sigstore)No build attestation published for this releaseNon vérifié
- Licence open source (OSI)MITspdx.org/licenses/MIT.html ↗Réussi
- Commit ou version publiée au cours de l'année écouléeLast commit or release 2026-08-23github.com/threadlinqs-cmd/intelthreadlinqs-mcp ↗Réussi
- OpenSSF ScorecardNon vérifié
- Nom du package non proche d'un package populaireNo popular package one edit away from intelthreadlinqs-mcpRéussi
- Aucun avis critique dans le package ou ses dépendances99 resolved dependencies checked; no critical advisoryosv.dev ↗Réussi
- Aucune instruction cachée destinée au modèle3 tool descriptions and the README read; no override phrases, credential paths or hidden charactersRéussi
- Les outils destructifs sont étiquetésNo tool named like delete, drop or execRéussi
- Le point de terminaison distant répond à la poignée de main MCPNon vérifié
README
Voir sur GitHubintelthreadlinqs-mcp
MCP server for Threadlinqs Intelligence — 81 tools and 25 analyst-playbook prompts across threat intelligence, detections, IOCs, threat actors, MITRE attack-chains, C2 infrastructure, and Purple-tier composite intelligence. Drop-in for Claude Code, Claude Desktop, Cursor, and any MCP-compatible client.
What this gives your AI agent
- 1,700+ threats with full metadata, severity, attribution, CVE/CWE, MITRE
- 16,000+ detection rules in Splunk SPL, Microsoft KQL, and Sigma YAML
- 47,000+ IOCs (IPs, domains, hashes, URLs, behavioral)
- 500+ threat actors with TTPs profiles and cross-actor infrastructure links
- 140+ live C2 beacons (Cobalt Strike) with watermark clustering + operator attribution
- Daily intelligence debriefs + 7-engine correlation analytics
- MITRE attack-chain prediction based on 4,271 observed technique transitions
Quick install
# No install needed — npx will fetch it
npx -y intelthreadlinqs-mcp
Claude Code
claude mcp add threadlinqs-intel \
-e THREADLINQS_API_KEY=tl_your_key_here \
-- npx -y intelthreadlinqs-mcp
The -e THREADLINQS_API_KEY is required in practice. The Threadlinqs Intelligence MCP server is a Purple-tier feature — it checks your key's tier at startup and warns if it is missing, invalid, or below Purple. The server still starts and exposes its tool catalog so clients and registries can introspect it, but tool calls are gated server-side: without a Purple or Gold key (tier ≥ 3) they return an Access denied error instead of data. There is no free or anonymous access to gated intelligence.
Claude Desktop
~/Library/Application Support/Claude/claude_desktop_config.json:
{
"mcpServers": {
"threadlinqs-intel": {
"command": "npx",
"args": ["-y", "intelthreadlinqs-mcp"],
"env": {
"THREADLINQS_API_KEY": "tl_your_key_here"
}
}
}
}
Cursor
.cursor/mcp.json:
{
"mcpServers": {
"threadlinqs-intel": {
"command": "npx",
"args": ["-y", "intelthreadlinqs-mcp"],
"env": {
"THREADLINQS_API_KEY": "tl_your_key_here"
}
}
}
}
VS Code
.vscode/mcp.json:
{
"servers": {
"threadlinqs-intel": {
"type": "stdio",
"command": "npx",
"args": ["-y", "intelthreadlinqs-mcp"],
"env": {
"THREADLINQS_API_KEY": "tl_your_key_here"
}
}
}
}
Getting an API key
Sign up at intel.threadlinqs.com, verify your email, and head to Profile → API Key. Note that signing up alone lands you on Blue (tier 1) — MCP tool calls need tier ≥ 3, so they return an under-tier error until you upgrade. A 7-day Purple free trial is offered at checkout on the pricing page.
Access — Purple tier only
The MCP server is a Purple-tier feature: all 81 tools require a Purple or Gold subscription (tier ≥ 3). The server checks your key's tier at startup and warns when it is missing, invalid, or below Purple — it still starts and lists its tools so clients can introspect the catalog, but the tier is enforced server-side on every call, so gated tools return Access denied rather than data. There is no free or anonymous access to gated intelligence.
| Tier | Price | MCP access |
|---|---|---|
| Purple | $11.99/mo | ✅ All 81 tools |
| Gold | Custom | ✅ All 81 tools (team — contact sales) |
| Lower tiers | — | ❌ No MCP access (the public website + REST API keep their own free Blue tier) |
A 7-day Purple free trial is available at checkout on the pricing page — it is not granted by signup alone. Tool calls enforce the tier server-side and return a structured 403 if your subscription lapses.
v4.3 flagship tools (Purple tier)
The composite tools are the reason most people upgrade to Purple — each one replaces 5–7 single-purpose MCP calls.
get_threat_hunting_bundle ⭐
Input: threat_id (e.g. "TL-2026-0599")
Returns: complete hunt dossier in one shot — threat metadata, full IOC list, SPL/KQL/Sigma detection queries, similar threats, simulation commands, and cross-threat infrastructure pivots. The single most useful tool in the platform.
get_actor_intelligence
Input: actor name (e.g. "Lazarus Group", "APT29")
Returns: comprehensive adversary picture — actor profile, attributed threats, MITRE techniques, IOCs (200 cap), detection rules (100 cap), activity timeline, active C2 infrastructure correlated to the actor, and cross-actor shared entities.
get_ioc_intelligence
Input: ioc_value (IP, domain, hash, URL)
Returns: every threat that touches the IOC + actor attribution + DNS enrichment trail + cross-IOC infrastructure pivots + consensus confidence score across 7 external feeds (Pulsedive, GreyNoise, YARAify, MalwareBazaar, URLScan, VxVault, OpenPhish). The "I found this in a log — tell me everything" workflow.
get_cve_intelligence
Input: cve_id (e.g. "CVE-2024-3400")
Returns: CVE detail + linked threats + EPSS exploitation velocity + KEV status + detection coverage % + available attack simulations + first-weaponization timeline.
get_mitre_gap_analysis
Input: optional tactic filter, limit
Returns: prioritized list of MITRE techniques without detection coverage, sorted by debt score (threat exposure + KEV count + EPSS). Each entry includes example threats and recommended detection types. Answers "what should I write detections for next?"
predict_mitre_transitions
Input: technique_id (e.g. "T1566"), top_n, direction (forward | reverse)
Returns: ranked next-technique predictions with probability and observation count, plus example threats showing the chain. Built from 4,271 observed transitions across the corpus.
generate_c2_blocklist
Input: none.
Returns: firewall-ready blocklist of active C2 IPs with country, ASN, version, watermark, and last-seen metadata. Currently tracking Cobalt Strike beacons. For operator clustering call get_c2, and for unmasked C2 domains get_c2_dns_intel.
Specialized tools
search_actors— Find threat actors by name, alias, nation-state, or motivation.get_actor— Full actor dossier in a single call.get_similar_threats— Precomputed-similarity matches by shared TTPs, IOC overlap, and same-actor attribution.
Agent skill
A portable skill that teaches any MCP client how to route across these tools — which tool to
call first, the hunt query grammar, the response budget, and the data caveats that keep
answers honest. Generated from the live catalog, so its tool reference cannot drift.
- Single file (paste into any assistant): https://intel.threadlinqs.com/mcp/skill.md
- Bundle (
SKILL.md+references/, for Claude Desktop / API skill upload): https://intel.threadlinqs.com/skills/intelthreadlinqs-mcp-skill.zip - Also in this repo under
skill/, and linked from https://intel.threadlinqs.com/mcp.
Prompts — 25 analyst playbooks
Prompts are pre-built workflows that orchestrate the tools for you. Rather than working out which of 81 tools to chain, invoke a playbook and the agent gets an ordered, argument-correct plan. Between them they reach every tool on the server, so an agent that only reads prompts/list still finds the whole surface.
| Playbook | What it does |
|---|---|
orient |
Start here — catalog, corpus shape, categories, TLQL grammar |
triage_cve |
CVE end-to-end: severity, exploitation, exposure, remediation |
profile_actor |
Actor dossier: TTPs, targeting, attribution confidence |
hunt_ioc |
Indicator → linked threats, blast radius, blocking action |
hunt_corpus |
Aggregate questions in TLQL over the hunt index |
research_question |
Open-ended research via semantic corpus search |
map_detections_to_mitre |
Map a threat's SPL/KQL/Sigma to ATT&CK, flag gaps |
review_detection_gaps |
Highest-value detection debt, ranked by exposure |
write_detection |
Draft a deployable rule grounded in corpus logic |
predict_next_move |
Forecast the next (or preceding) ATT&CK techniques |
explain_link |
Why two threats are linked, with the evidence |
map_campaign |
Campaign members, shared infra, pivotal entities |
review_attribution |
Audit whether an attribution is evidence-backed |
pivot_infrastructure |
Pivot across DNS, hosting and adjacent infrastructure |
malware_dossier |
Profile a malware family or offensive tool |
build_c2_blocklist |
Firewall-ready C2 blocklist with operator clustering |
osint_sweep |
Fold community signal in; surface coverage lead time |
sweep_vulnerabilities |
Triage by exploitability, not CVSS alone |
plan_purple_team |
Simulations paired with the detections they should trip |
export_for_tooling |
STIX 2.1, ATT&CK Navigator, SIEM-ready rules |
bulk_enrich |
Batch a list of threat/CVE ids without N round-trips |
assess_exposure |
Covered vs uncovered for a CVE, actor or technique |
daily_brief |
Today's posture, top new threats, daily theme |
period_review |
Review a day or span from the debrief archive |
platform_status |
Corpus health, engine state, enrichment completeness |
Full tool catalog
For the complete list of 81 tools with parameters and example invocations, see the interactive MCP documentation page.
Architecture
-
Transport: stdio (local subprocess)
-
Auth:
THREADLINQS_API_KEYenvironment variable (Bearer token to the worker API) -
Runtime: Node ≥18
-
SDK:
@modelcontextprotocol/sdk@^1.26.0 -
Backend: Cloudflare Workers + D1 (multi-region)
-
Purple gate: the server checks the API key's tier at startup and warns if it is missing, invalid, or below Purple/Gold (tier ≥ 3) — it starts regardless so the tool catalog stays introspectable; the gate itself is enforced server-side on every tool call (structured 403 on lapse or under-tier)
-
Introspection has three tiers (since 8.1.0): authenticated
tools/list→ the public/mcp/catalog.json→ a catalog snapshot bundled in the package. The first two need network; the third does not, sotools/listreturns the real catalog even in a fully network-isolated sandbox. Previously that case returned an empty array — not an error, an empty list, which is indistinguishable from a server with no tools, and is why registry scanners indexed this server with zero tools.Maintainers: regenerate the snapshot with
npm run sync:catalogafter deploying the worker, since the worker owns the tool and prompt registries.npm run test:catalogfails the build when the snapshot and the live catalog disagree, and it runs onprepublishOnly.
Links
License
MIT © Threadlinqs
Questions sur Threadlinqs Intelligence MCP
Comment ajouter Threadlinqs Intelligence MCP à Claude Code ?
Exécutez claude mcp add --env 'THREADLINQS_API_KEY=<YOUR_THREADLINQS_API_KEY>' intelthreadlinqs-mcp -- npx -y intelthreadlinqs-mcp dans votre terminal. Ajoutez --scope user pour le rendre disponible dans tous les projets.
Comment ajouter Threadlinqs Intelligence MCP à Cursor ?
Ouvrez ~/.cursor/mcp.json et ajoutez le serveur sous mcpServers, avec la configuration Cursor indiquée dans la section d'installation ci-dessus.
Quels outils Threadlinqs Intelligence MCP fournit-il ?
Il expose 3 outils, dont search_actors, get_actor, get_similar_threats.
Existe-t-il une offre gratuite ?
L'inscription donne accès au niveau Blue, mais les appels d'outils MCP nécessitent un abonnement Purple ou Gold. Un essai gratuit de 7 jours pour le niveau Purple est disponible lors du paiement sur la page des tarifs.
Quel type de règles de détection le serveur propose-t-il ?
Le serveur inclut plus de 16 000 règles de détection formatées en Splunk SPL, Microsoft KQL et Sigma YAML.