AI Sec Watch: A Security Intelligence Platform for AI Systems

Luu, T.J.

Public API

Access AI Sec Watch data programmatically. All endpoints return JSON with CORS enabled, so you can call them from any domain.

OpenAPI SpecBase URL: /api/v1Rate limit: 60 req/min per IPCORS: * (all origins)

Quick Start

import pandas as pd

# Load full dataset (CSV)
df = pd.read_csv("https://aisecwatch.com/api/issues/export")

# Or use the JSON API with filtering
import requests

resp = requests.get("https://aisecwatch.com/api/v1/issues", params={
    "severity": "critical",
    "limit": 100,
})
data = resp.json()["data"]

No API key required. 60 requests/minute. All data CC-BY-4.0.

GET/api/v1/issues

List relevant, processed security issues with optional filtering and pagination.

Query Parameters

searchstringSearch issues by title, summary, and CVE ID

labelstringFilter by label (e.g. "security", "privacy", "research", "safety")

severitystringFilter by severity (e.g. "critical", "high", "medium", "low")

issueTypestringFilter by issue type: "vulnerability", "incident", "news", "regulatory", "research"

vendorstringFilter by affected vendor (e.g. "OpenAI", "Anthropic", "Google", "LangChain")

sinceISO 8601 dateOnly return issues published after this date

limitnumberResults per page (default 20, max 100)

offsetnumberNumber of results to skip (default 0)

confidenceMinnumberMinimum classifier confidence 0-1 (e.g. 0.8 for >80%)

Example Response

{
  "data": [
    {
      "id": "a1b2c3d4-...",
      "title": "Critical prompt injection in LangChain",
      "summary": "A critical vulnerability was found...",
      "sourceUrl": "https://...",
      "publishedAt": "2026-02-10T12:00:00.000Z",
      "severity": "critical",
      "cvssSeverity": "CRITICAL",
      "cvssScore": "9.8",
      "labels": ["security"],
      "issueType": "vulnerability",
      "cveId": "CVE-2026-1234",
      "cweIds": ["CWE-77"],
      "affectedPackages": ["langchain"],
      "affectedVendors": ["LangChain"],
      "affectedVendorsRaw": ["LangChain", "OpenAI"],
      "classifierModel": "claude-haiku-4-5-20251001",
      "classifierPromptVersion": "v3",
      "solution": "Upgrade to v0.1.1 or later",
      "attackType": ["prompt_injection"],
      "cvssVector": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H",
      "attackVector": "network",
      "attackComplexity": "low",
      "privilegesRequired": "none",
      "userInteraction": "none",
      "exploitMaturity": "active",
      "epssScore": 0.87,
      "patchAvailable": true,
      "disclosureDate": "2026-02-08T00:00:00.000Z",
      "capecIds": ["CAPEC-242"],
      "crossRefCount": 3,
      "attackSophistication": "moderate",
      "impactType": ["confidentiality", "integrity"],
      "aiComponentTargeted": "framework",
      "llmSpecific": true,
      "classifierConfidence": 0.92
    }
  ],
  "meta": {
    "total": 142,
    "limit": 20,
    "offset": 0
  }
}

GET/api/v1/issues/:id

Get a single issue by its UUID. Returns 404 if not found or not relevant.

Example Response

{
  "data": {
    "id": "a1b2c3d4-...",
    "title": "Critical prompt injection in LangChain",
    "summary": "A critical vulnerability was found...",
    "sourceUrl": "https://...",
    "publishedAt": "2026-02-10T12:00:00.000Z",
    "severity": "critical",
    "cvssSeverity": "CRITICAL",
    "cvssScore": "9.8",
    "labels": ["security"],
    "issueType": "vulnerability",
    "cveId": "CVE-2026-1234",
    "cweIds": ["CWE-77"],
    "affectedPackages": ["langchain"],
    "solution": "Upgrade to v0.1.1 or later",
    "attackType": ["prompt-injection"]
  }
}

GET/api/v1/digest/latest

Get the latest daily TLDR digest. Falls back to the most recent non-empty digest.

Example Response

{
  "data": {
    "date": "2026-02-12",
    "content": "## AI Security TLDR — Feb 12, 2026\n\n...",
    "issueCount": 8,
    "generatedAt": "2026-02-12T06:00:00.000Z"
  }
}

GET/api/v1/issues/stix

Export issues as a STIX 2.1 bundle. Compatible with MISP, OpenCTI, ThreatConnect, and other threat intelligence platforms.

Query Parameters

severitystringFilter by severity (e.g. "critical", "high")

issueTypestringFilter by issue type (e.g. "vulnerability")

sinceISO 8601 dateOnly return issues published after this date

limitnumberMax issues to include (default 50, max 100)

Example Response

{
  "type": "bundle",
  "id": "bundle--a1b2c3d4-...",
  "objects": [
    {
      "type": "identity",
      "spec_version": "2.1",
      "id": "identity--d3c1e7a0-...",
      "name": "AI Sec Watch",
      "identity_class": "system"
    },
    {
      "type": "vulnerability",
      "spec_version": "2.1",
      "id": "vulnerability--a1b2c3d4-...",
      "name": "CVE-2026-1234",
      "description": "Critical prompt injection...",
      "external_references": [
        { "source_name": "cve", "external_id": "CVE-2026-1234" }
      ]
    },
    {
      "type": "attack-pattern",
      "spec_version": "2.1",
      "id": "attack-pattern--e5f6a7b8-...",
      "name": "Prompt Injection"
    }
  ]
}

GET/api/v1/stats

Get aggregate platform statistics.

Example Response

{
  "data": {
    "totalIssues": 1420,
    "totalDigests": 85,
    "last24h": 12,
    "last7d": 67
  }
}

POST/api/v1/webhooks

Create a webhook to receive real-time notifications when new issues match your filters. Requires WEBHOOK_API_KEY bearer auth. Payloads are signed with HMAC-SHA256.

Query Parameters

urlstring (body)The HTTPS URL to receive webhook POSTs

filtersobject (body)Optional filters: { severity, issueType, vendor, llmSpecific }

Example Response

// Request
POST /api/v1/webhooks
Authorization: Bearer YOUR_WEBHOOK_API_KEY
{
  "url": "https://your-server.com/webhook",
  "filters": { "severity": "critical", "llmSpecific": true }
}

// Response (201)
{
  "data": {
    "id": "a1b2c3d4-...",
    "url": "https://your-server.com/webhook",
    "secret": "hmac_secret_hex_string",
    "active": true
  }
}

// Webhook payload (POST to your URL)
{
  "event": "new_issues",
  "issues": [ ... ],
  "deliveredAt": "2026-03-12T12:00:00.000Z"
}
// Header: X-Webhook-Signature: sha256=...

Error Responses

All errors follow a consistent format:

{
  "error": {
    "message": "Rate limit exceeded. Max 60 requests per minute.",
    "status": 429
  }
}

400 — Bad request (invalid parameters)

404 — Resource not found

429 — Rate limit exceeded

500 — Internal server error

How to Cite

If you use this dataset in academic research, industry analysis, or threat intelligence work, please cite it. You can copy the citation text or download it as a file for import into your reference manager (Zotero, Mendeley, EndNote, etc.).

@misc{luu2026aisecwatch,
  author       = {Luu, T.J.},
  title        = {{AI Sec Watch Dataset}: Structured {AI} Security Threat Intelligence},
  year         = {2026},
  howpublished = {\url{https://aisecwatch.com/api-docs}},
  note         = {Accessed: March 17, 2026}
}

Quick Start

import pandas as pd

# Load full dataset (CSV)
df = pd.read_csv("https://aisecwatch.com/api/issues/export")

# Or use the JSON API with filtering
import requests

resp = requests.get("https://aisecwatch.com/api/v1/issues", params={
    "severity": "critical",
    "limit": 100,
})
data = resp.json()["data"]

No API key required. 60 requests/minute. All data CC-BY-4.0.

{ "data": [ { "id": "a1b2c3d4-...", "title": "Critical prompt injection in LangChain", "summary": "A critical vulnerability was found...", "sourceUrl": "https://...", "publishedAt": "2026-02-10T12:00:00.000Z", "severity": "critical", "cvssSeverity": "CRITICAL", "cvssScore": "9.8", "labels": ["security"], "issueType": "vulnerability", "cveId": "CVE-2026-1234", "cweIds": ["CWE-77"], "affectedPackages": ["langchain"], "affectedVendors": ["LangChain"], "affectedVendorsRaw": ["LangChain", "OpenAI"], "classifierModel": "claude-haiku-4-5-20251001", "classifierPromptVersion": "v3", "solution": "Upgrade to v0.1.1 or later", "attackType": ["prompt_injection"], "cvssVector": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H", "attackVector": "network", "attackComplexity": "low", "privilegesRequired": "none", "userInteraction": "none", "exploitMaturity": "active", "epssScore": 0.87, "patchAvailable": true, "disclosureDate": "2026-02-08T00:00:00.000Z", "capecIds": ["CAPEC-242"], "crossRefCount": 3, "attackSophistication": "moderate", "impactType": ["confidentiality", "integrity"], "aiComponentTargeted": "framework", "llmSpecific": true, "classifierConfidence": 0.92 } ], "meta": { "total": 142, "limit": 20, "offset": 0 } }

{ "data": { "id": "a1b2c3d4-...", "title": "Critical prompt injection in LangChain", "summary": "A critical vulnerability was found...", "sourceUrl": "https://...", "publishedAt": "2026-02-10T12:00:00.000Z", "severity": "critical", "cvssSeverity": "CRITICAL", "cvssScore": "9.8", "labels": ["security"], "issueType": "vulnerability", "cveId": "CVE-2026-1234", "cweIds": ["CWE-77"], "affectedPackages": ["langchain"], "solution": "Upgrade to v0.1.1 or later", "attackType": ["prompt-injection"] } }

{ "type": "bundle", "id": "bundle--a1b2c3d4-...", "objects": [ { "type": "identity", "spec_version": "2.1", "id": "identity--d3c1e7a0-...", "name": "AI Sec Watch", "identity_class": "system" }, { "type": "vulnerability", "spec_version": "2.1", "id": "vulnerability--a1b2c3d4-...", "name": "CVE-2026-1234", "description": "Critical prompt injection...", "external_references": [ { "source_name": "cve", "external_id": "CVE-2026-1234" } ] }, { "type": "attack-pattern", "spec_version": "2.1", "id": "attack-pattern--e5f6a7b8-...", "name": "Prompt Injection" } ] }

// Request POST /api/v1/webhooks Authorization: Bearer YOUR_WEBHOOK_API_KEY { "url": "https://your-server.com/webhook", "filters": { "severity": "critical", "llmSpecific": true } } // Response (201) { "data": { "id": "a1b2c3d4-...", "url": "https://your-server.com/webhook", "secret": "hmac_secret_hex_string", "active": true } } // Webhook payload (POST to your URL) { "event": "new_issues", "issues": [ ... ], "deliveredAt": "2026-03-12T12:00:00.000Z" } // Header: X-Webhook-Signature: sha256=...

How to Cite

@misc{luu2026aisecwatch, author = {Luu, T.J.}, title = {{AI Sec Watch Dataset}: Structured {AI} Security Threat Intelligence}, year = {2026}, howpublished = {\url{https://aisecwatch.com/api-docs}}, note = {Accessed: March 17, 2026} }