# CVE Security Advisory Triage (`jetlaunch/cve-security-triage`) Actor Rank CVEs and package vulnerabilities by priority using NVD, CISA KEV, EPSS and OSV, with a plain verdict and the reasons behind it. - **URL**: https://apify.com/jetlaunch/cve-security-triage.md - **Developed by:** [Chris O'Byrne](https://apify.com/jetlaunch) (community) - **Categories:** Developer tools, Automation - **Stats:** 2 total users, 1 monthly users, 100.0% runs succeeded, 0 bookmarks - **User rating**: No ratings yet ## Pricing Pay per event This Actor is paid per event. You are not charged for the Apify platform usage, but only a fixed price for specific events. Learn more: https://docs.apify.com/actors/running/actors-in-store.md#pay-per-event ## What's an Apify Actor? An Actor is a serverless cloud program that runs on the Apify platform. It has two run modes. In Batch mode, an Actor accepts a well-defined JSON input, performs an action which can take anything from a few seconds to a few hours, and optionally produces a well-defined JSON output, datasets with results, or files in key-value store. In Standby mode, an Actor provides a web server which can be used as a website, API, or an MCP server. Apify vocabulary and the platform model are defined once, in the agent quickstart at https://apify.com/agents.md. ## How to integrate an Actor? If asked about integration, you help developers integrate Actors into their projects. You adapt to their stack and deliver integrations that are safe, well-documented, and production-ready. Do not guess an integration path. Every one of them is in the agent quickstart at https://apify.com/agents.md: the Apify MCP server, Agent Skills with the Apify CLI, the JavaScript and Python clients, the REST API, and the account-free path for an agent with no human to sign in. It also carries the rule on stating cost before the first paid run. For examples already wired to this Actor's own input schema, see the [API](#api) section below. Each client library has reference documentation the quickstart does not restate: [JavaScript/TypeScript](https://docs.apify.com/api/client/js/docs.md) (`npm install apify-client`) and [Python](https://docs.apify.com/api/client/python/docs.md) (`pip install apify-client`). # README ### CVE Security Advisory Triage CVE Security Advisory Triage turns a list of CVE ids, a keyword or a package watchlist into a ranked, plain-language priority list. For each vulnerability it combines the **CVSS v3 and v4 scores** from the NVD, the **CISA Known Exploited Vulnerabilities (KEV)** status and due date, the **EPSS exploit probability** from FIRST, and **affected packages with fixed-in versions** from OSV. It then gives a verdict (Act now, High, Medium, Low or Unscored) and lists the reasons behind it, so you can see why an item ranks where it does. The easiest way to try it: open the Input tab and click **Start**. The prefilled input triages CVE-2021-44228 (Log4Shell) and CVE-2024-3094 (the xz backdoor). **This Actor is decision support, not a guarantee of safety.** The verdict is a rule-based summary of public data. A low verdict does not mean a vulnerability cannot be exploited, and the data sources can be incomplete or late. Always check vendor advisories and apply your own judgement. #### Who is this for? - Security and platform teams who need to decide which patches come first - Developers checking the dependencies of an application or a container image - Consultants and auditors who want a repeatable, sourced prioritisation list - Anyone feeding vulnerability data into tickets, dashboards or alerts through the Apify API #### Two modes - **cve-triage**: give CVE ids (up to 100) or a keyword. You get one row per CVE with scores, KEV and EPSS fields, affected packages, references and a verdict. - **watchlist-scan**: give up to 50 packages (name, ecosystem, version) or complete CPE 2.3 strings. You get every known vulnerability for those versions, ranked by priority, with the best-effort fixed version for the version you run. #### How to use it 1. Choose the mode. 2. For cve-triage, enter CVE ids or a keyword. For watchlist-scan, paste a JSON list of packages, for example `{"name": "lodash", "ecosystem": "npm", "version": "4.17.15"}`. 3. Optional: add your own free NVD API key to make cve-triage runs faster (see FAQ). 4. Start the run and open the Output tab. Use the table views or export JSON, CSV or Excel. 5. Schedule the Actor or call it through the API to repeat the check daily. #### Input | Field | Mode | Description | | --- | --- | --- | | mode | both | `cve-triage` or `watchlist-scan` | | cveIds | cve-triage | CVE ids, up to 100 | | keyword | cve-triage | NVD description search, ranked by priority after enrichment | | maxKeywordResults | cve-triage | 1 to 50 CVEs taken from the keyword search (default 10) | | keywordExactMatch | cve-triage | Match the keyword as an exact phrase | | packages | watchlist-scan | Objects with name, ecosystem, version. OSV ecosystem names such as npm, PyPI, Maven, Go, crates.io, NuGet, RubyGems, Packagist | | cpes | watchlist-scan | Complete CPE 2.3 names, up to 20 (packages plus CPEs must total 50 or fewer) | | maxVulnsPerPackage | watchlist-scan | Advisories loaded per package, default 150 | | minimumVerdict | watchlist-scan | Hide findings below Medium, High or Act now | | nvdApiKey | both | Optional secret. Your own NVD API key | #### Output Results are stored in the default dataset, with a table view for each mode. A `SUMMARY` record in the key-value store lists counts, the KEV catalog version used, items that could not be resolved, and the attribution text. Example row from cve-triage (trimmed): ```json { "mode": "cve-triage", "cve_id": "CVE-2023-4863", "verdict": "Act now", "priority_score": 95, "reasons": [ "Listed in the CISA Known Exploited Vulnerabilities catalog (added 2023-09-13, federal remediation due 2023-10-04).", "EPSS 1.000 (100.0th percentile): estimated probability of exploitation activity in the next 30 days.", "CVSS v3.1 8.8 (HIGH) describes technical severity.", "NVD references include an item tagged Exploit (a public exploit or proof of concept may exist).", "A fixed version is documented, for example google/chrome@116.0.5845.187, mozilla/firefox@102.15.1, mozilla/firefox@117.0.1." ], "recommended_action": "Patch or mitigate first. CISA required action: Apply mitigations per v...", "severity": "HIGH", "cvss_v3_score": 8.8, "cvss_v3_severity": "HIGH", "cvss_v3_vector": "CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:U/C:H/I:H/A:H", "cvss_v3_source": "nvd@nist.gov", "cvss_v4_score": null, "cvss_v4_severity": null, "cvss_v4_vector": null, "cvss_v2_score": null, "kev_listed": true, "kev_cve_id": "CVE-2023-4863", "kev_date_added": "2023-09-13", "kev_due_date": "2023-10-04", "kev_required_action": "Apply mitigations per vendor instructions or discontinue use of the pr...", "kev_known_ransomware_use": "Unknown", "kev_vendor_project": "Google", "kev_product": "Chromium WebP", "epss_score": 0.99979, "epss_percentile": 0.99981, "epss_date": "2026-10-21", "description": "Heap buffer overflow in libwebp in Google Chrome prior to 116.0.5845.187 and libwebp 1.3.2 allowed a remote attacker to ...", "nvd_status": "Analyzed", "published": "2023-09-12T15:15:24.327", "last_modified": "2026-06-17T06:38:46.547", "cwe_ids": [ "CWE-787" ], "fixed_in": [ "google/chrome@116.0.5845.187", "mozilla/firefox@102.15.1", "mozilla/firefox@117.0.1", "..." ], "affected_packages": [ { "source": "osv", "ecosystem": "crates.io", "name": "libwebp-sys2", "ranges": [ { "introduced": "0", "fixed": "0.1.8", "lastAffected": null } ], "fixedIn": [ "0.1.8" ] }, { "source": "nvd-cpe", "ecosystem": null, "name": "google/chrome", "cpe": "cpe:2.3:a:google:chrome:*:*:*:*:*:*:*:*", "affectedVersion": null, "versionRange": { "startIncluding": null, "startExcluding": null, "endIncluding": null, "endExcluding": "116.0.5845.187" }, "fixedIn": [ "116.0.5845.187" ] } ], "exploit_reference_present": true, "references": [ { "url": "https://github.com/webmproject/libwebp/commit/902bc9190331343b2017211debcec8d2ab87e17a", "tags": [ "Patch" ] } ], "reference_count": 47, "data_quality": "complete", "data_notes": [], "sources": { "nvd": "ok", "kev": "ok", "epss": "ok", "osv": "ok" }, "retrieved_at": "2026-10-21T20:55:30.000Z" } ``` In watchlist-scan mode each row has `rank`, `package`, `ecosystem`, `installed_version`, `vuln_id`, `cve_ids`, `verdict`, `priority_score`, the same CVSS, KEV and EPSS fields, `fixed_in` (best-effort fixed version for your installed version), `fixed_versions_all` and `references`. #### How the verdict works | Verdict | Rule (first match wins) | | --- | --- | | Act now | Listed in CISA KEV, or EPSS of 0.5 or more, or an OSV malicious package advisory | | High | EPSS of 0.1 or more, or CVSS 9.0 or more, or CVSS 7.0 or more with EPSS of 0.02 or more | | Medium | CVSS 7.0 or more, or EPSS of 0.01 or more, or CVSS 4.0 or more | | Low | Scored, but below the Medium thresholds | | Unscored | No CVSS and no EPSS data yet | | Rejected | NVD marks the CVE as Rejected | Every row lists the signals used in `reasons`. `priority_score` (0 to 100) only orders rows inside the same verdict. It adds up to 40 points for CVSS, up to 40 for EPSS (full marks at 0.5) and 20 for KEV listing. The thresholds are a transparent starting point and not an official standard. #### Pricing This Actor uses pay-per-event pricing. You pay only for the events below, with no separate compute charge. | Event | Price | When it is charged | | --- | --- | --- | | cve-triage | $0.01 per CVE | Once per CVE returned with a complete NVD record | | watchlist-scan | $0.05 per scan | Once per scan of up to 50 packages or CPE strings that returns at least one finding | Nothing is charged for failed lookups, CVEs that NVD does not know, partial results marked `data_quality: partial`, invalid input, or scans with no findings. Examples: 10 CVEs cost $0.10, 100 CVEs cost $1.00, and a 50-package scan costs $0.05 however many findings it returns. If you set a maximum cost per run, the Actor stops before it exceeds it. #### Data sources, licences and attribution - **NVD** (NIST National Vulnerability Database), CVE API 2.0. This product uses the NVD API but is not endorsed or certified by the NVD. Each CVE is fetched from the NVD on demand, one request at a time, and the Actor keeps to the published unauthenticated limit of 5 requests per rolling 30 seconds unless you supply your own key. - **CISA Known Exploited Vulnerabilities catalog**, published by CISA under a CC0 licence. - **EPSS** scores are provided by FIRST. See EPSS at https://www.first.org/epss - **OSV** (https://osv.dev), an aggregate of advisories from the GitHub Advisory Database, PyPA, Go, RustSec and others, each under its own licence (mostly CC-BY 4.0 or CC0). The KEV catalog is downloaded at most once per day, and each CVE's EPSS score is requested at most once per day. Both are cached in a named key-value store called `cve-triage-cache` in your own Apify account. This Actor is not affiliated with or endorsed by NIST, the NVD, CISA, FIRST or the OSV project. #### FAQ **How long does a run take?** Without an NVD key, about 6.5 seconds per CVE, because the Actor spaces NVD requests to stay inside the unauthenticated limit. In our test 5 CVEs took 31 seconds, and 100 CVEs should take roughly 11 minutes. A watchlist-scan of 50 packages took 26 seconds in our test and returned 640 findings, because package queries use OSV and not the NVD. **Do I need an NVD API key?** No. A free personal key from nvd.nist.gov raises the NVD limit to 50 requests per 30 seconds. Enter it in the secret field. The key is only used for requests to the NVD, and the NVD asks that a key is used only by the person who requested it. **Why is a CVE marked partial?** The NVD could not be reached, or has no record yet, but KEV or EPSS data exists. You still get the KEV and EPSS fields, and the row is not charged. **Why is fixed_in empty?** None of the sources checked lists a fixed version. For watchlist-scan, `fixed_in` is the smallest fixed version above your installed version found by a generic version comparison, so check it against the vendor advisory. **How does this differ from NVD search Actors or CVE-to-exploit mappers?** Those return NVD records or links to public proof-of-concept code. This Actor adds KEV, EPSS and OSV fix data and returns one ranked verdict with reasons. It does not search for or return exploit code. **Can I use the API?** Yes. Run it with the Apify API, schedule it, or connect it to integrations such as webhooks, and read the dataset in JSON, CSV or Excel. #### Limitations - NVD limits are per IP address, and Apify runs share IP addresses. If the NVD answers with rate-limit errors the Actor waits and retries, and a very busy period can slow a run or produce partial rows. Your own NVD key avoids most of this. - Keyword results come from the NVD in its own order. Only the first `maxKeywordResults` CVEs are ranked. - watchlist-scan reads CVSS from OSV vectors. CVSS v4 is shown as a vector only. It does not call the NVD for package queries, so it can lag the NVD for very new CVEs. CPE queries return the first 200 CVEs per CPE. - Advisories without a CVE id have no KEV or EPSS data. - Data can be wrong, late or missing. This is not a vulnerability scanner of your systems. #### Support Questions, bug reports and feature requests are welcome in the Issues tab of this Actor. Include the run ID and the CVE ids or packages involved. # Changelog This Actor's version history is a separate document: https://apify.com/jetlaunch/cve-security-triage/changelog.md # Actor input Schema ## `mode` (type: `string`): cve-triage looks up CVE ids (or a keyword) and returns scores, KEV and EPSS status, affected packages and a priority verdict. watchlist-scan checks up to 50 packages or CPE strings and returns known vulnerabilities ranked by priority. ## `cveIds` (type: `array`): CVE ids such as CVE-2021-44228. Up to 100 per run. Each CVE is looked up in NVD one at a time, so a run takes about 7 seconds per CVE without an NVD API key. ## `keyword` (type: `string`): Optional. Searches NVD descriptions, for example "openssl heap overflow". Results are ranked by priority after enrichment. Used together with the CVE ids above, or on its own. ## `maxKeywordResults` (type: `integer`): How many CVEs to take from a keyword search (1 to 50). Each CVE returned is billed as one cve-triage event. ## `keywordExactMatch` (type: `boolean`): Match the keyword as an exact phrase instead of all words anywhere in the description. ## `packages` (type: `array`): List of objects with name, ecosystem and version, for example {"name": "lodash", "ecosystem": "npm", "version": "4.17.15"}. Ecosystems follow OSV names: npm, PyPI, Maven, Go, crates.io, NuGet, RubyGems, Packagist, Hex, Pub and more. Up to 50 packages plus CPE strings per scan. ## `cpes` (type: `array`): Optional complete CPE 2.3 names, for example cpe:2.3:a:apache:log4j:2.14.1:*:*:*:*:*:*:\*. Each one is one NVD request (about 7 seconds without an API key). Up to 20, and the combined total with packages must be 50 or fewer. ## `maxVulnsPerPackage` (type: `integer`): Caps how many OSV advisories are loaded for each package (1 to 500). Advisories are not ranked before they are loaded, so keep this high enough to cover old versions. ## `minimumVerdict` (type: `string`): Hide findings below this priority. A scan is billed once if at least one finding is returned. ## `nvdApiKey` (type: `string`): Optional. Your own free NVD API key raises the NVD limit from 5 to 50 requests per 30 seconds, which makes cve-triage runs faster. Without a key the Actor keeps to the unauthenticated limit. Request a key at nvd.nist.gov/developers/request-an-api-key. ## Actor input object example ```json { "mode": "cve-triage", "cveIds": [ "CVE-2021-44228", "CVE-2024-3094" ], "maxKeywordResults": 10, "keywordExactMatch": false, "packages": [ { "name": "lodash", "ecosystem": "npm", "version": "4.17.15" }, { "name": "django", "ecosystem": "PyPI", "version": "3.2.0" } ], "maxVulnsPerPackage": 150, "minimumVerdict": "all" } ``` # Actor output Schema ## `results` (type: `string`): One row per CVE (cve-triage) or per finding (watchlist-scan). ## `summary` (type: `string`): Counts, data source versions, attribution and any CVEs that could not be resolved. # API You can run this Actor programmatically using our API. Below are code examples in JavaScript, Python, and CLI, as well as the OpenAPI specification and MCP server setup. ## JavaScript example ```javascript import { ApifyClient } from 'apify-client'; // Initialize the ApifyClient with your Apify API token // Replace the '' with your token const client = new ApifyClient({ token: '', }); // Prepare Actor input const input = { "cveIds": [ "CVE-2021-44228", "CVE-2024-3094" ], "packages": [ { "name": "lodash", "ecosystem": "npm", "version": "4.17.15" }, { "name": "django", "ecosystem": "PyPI", "version": "3.2.0" } ] }; // Run the Actor and wait for it to finish const run = await client.actor("jetlaunch/cve-security-triage").call(input); // Fetch and print Actor results from the run's dataset (if any) console.log('Results from dataset'); console.log(`💾 Check your data here: https://console.apify.com/storage/datasets/${run.defaultDatasetId}`); const { items } = await client.dataset(run.defaultDatasetId).listItems(); items.forEach((item) => { console.dir(item); }); // 📚 Want to learn more 📖? Go to → https://docs.apify.com/api/client/js/docs ``` ## Python example ```python from apify_client import ApifyClient # Initialize the ApifyClient with your Apify API token # Replace '' with your token. client = ApifyClient("") # Prepare the Actor input run_input = { "cveIds": [ "CVE-2021-44228", "CVE-2024-3094", ], "packages": [ { "name": "lodash", "ecosystem": "npm", "version": "4.17.15", }, { "name": "django", "ecosystem": "PyPI", "version": "3.2.0", }, ], } # Run the Actor and wait for it to finish run = client.actor("jetlaunch/cve-security-triage").call(run_input=run_input) # Fetch and print Actor results from the run's dataset (if there are any) print(f"💾 Check your data here: https://console.apify.com/storage/datasets/{run.default_dataset_id}") for item in client.dataset(run.default_dataset_id).iterate_items(): print(item) # 📚 Want to learn more 📖? Go to → https://docs.apify.com/api/client/python/docs/quick-start ``` ## CLI example ```bash echo '{ "cveIds": [ "CVE-2021-44228", "CVE-2024-3094" ], "packages": [ { "name": "lodash", "ecosystem": "npm", "version": "4.17.15" }, { "name": "django", "ecosystem": "PyPI", "version": "3.2.0" } ] }' | apify call jetlaunch/cve-security-triage --silent --output-dataset ``` ## MCP server setup ```json { "mcpServers": { "apify": { "type": "http", "url": "https://mcp.apify.com/?tools=fetch-actor-details,jetlaunch/cve-security-triage" } } } ``` The hosted server signs you in with OAuth on first connect, so no API token belongs in this config. Clients without OAuth support can send an `Authorization: Bearer ` header instead, using a token from API & Integrations in Apify Console (https://console.apify.com/settings/integrations). ## OpenAPI specification Download the OpenAPI definition: https://api.apify.com/v2/actors/xKEGAEhI8iQjWciZY/builds/H88V5vmUPALB5TTEp/openapi.json