Aikido Intel

Supply chain threat intelligence, honestly reviewed. No marketing fluff, just what you get when you use it.

TL;DR

• What it is: An AI-powered, open-source vulnerability and malware feed that monitors open-source packages across 12+ ecosystems for threats — including silently patched vulnerabilities that never reach CVE databases [1].
• Who it’s for: Developers, security engineers, and SaaS vendors who want earlier warning on supply chain threats than NVD, OSV, or GitHub Advisory Database can provide. Also relevant for teams building security products who want to license a threat intelligence data source [1][5].
• License: AGPL — you can use, modify, and distribute the feed freely. Commercial API access for product integration is a separate paid tier with pricing not publicly disclosed [homepage].
• Key strength: Catches vulnerabilities before they’re disclosed anywhere else. As of the last published count, 67% of the vulnerabilities Aikido Intel discovered were never publicly reported to any vulnerability database — including high and critical severity ones [1].
• Key weakness: Aikido Intel is a threat intelligence feed, not a full security scanner. It tells you what’s vulnerable in the open-source ecosystem, but integrating that knowledge into your stack requires either using the broader Aikido Security platform or licensing the API and building the plumbing yourself [homepage][3].
• GitHub stars: 179 — modest, but the product is a data feed, not a self-hosted application stack [GitHub].

---

What is Aikido Intel

Aikido Intel is a threat intelligence feed that watches open-source package registries — npm, PyPI, PHP/Packagist, Ruby, NuGet, Maven, Rust, Go, C++, Dart, Elixir, and Swift — and reports vulnerabilities and malware that haven’t been publicly disclosed yet.

The mechanism is straightforward: most open-source maintainers fix security issues without announcing them. They commit the patch, bump the version, and move on. No CVE request, no GitHub Security Advisory, no post. Your dependency scanner never sees it because scanners only flag issues with CVE numbers. You’re running vulnerable software and have no idea [1].

Aikido Intel uses custom-trained LLMs to read changelogs, release notes, and commit diffs across 4.3 million packages and identify when a security issue has been quietly fixed. It then cross-references five vulnerability databases. If there’s no match, a security engineer reviews and assigns an Aikido Vulnerability Number (AV-XXXX) and severity rating. The finding goes public immediately [1].

As of the December 2024 launch post, they had discovered 511 such vulnerabilities since launch in January 2024. The headline stat: 67% of the vulnerabilities they found were never publicly disclosed to any database. That includes over 50% of high and critical severity findings [1]. One concrete example from their blog: Axios, the HTTP client with 56 million weekly npm downloads and 146,000+ dependents, quietly patched a prototype pollution vulnerability in January 2024 that was never publicly reported [1].

The live feed at intel.aikido.dev shows the current count: ~2,000 tracked vulnerabilities and 125,200+ identified malware packages.

There’s a companion tool called Safe Chain (open source) that hooks into npm install to block malware from running during package installation. It’s listed separately but sits in the same threat intelligence product family [homepage].

---

Why people choose it over CVE-only feeds

The practical argument for Aikido Intel isn’t that it replaces NVD or OSV — it’s that it fills the gap those databases structurally can’t fill.

The silent patching problem is larger than people think. The 67% non-disclosure figure [1] is the stat that makes security engineers stop scrolling. Low-severity findings being silently patched is expected and arguably fine — maintainers have limited time and CVE requests add overhead. But Aikido’s data shows the same pattern in high and critical findings. That means any org relying exclusively on CVE-based scanning has a meaningful blind spot in their supply chain coverage [1][5].

Standard vulnerability feeds have a known lag. There’s typically days to weeks between a patch landing and a CVE being published. Aikido Intel claims detection “within minutes” of a release [homepage]. The comparison to watch for: a company using only NVD-based scanning might not flag a dependency risk until a CVE is assigned, long after the patched version has been available and the unpatched version is actively exploitable in the wild [5].

The malware detection angle is different. This isn’t about CVEs at all — it’s about packages that were published maliciously, typically via typosquatting, account hijacking, or dependency confusion. The 125,200+ malware packages in the feed cover backdoors, trojans, keyloggers, XSS injection scripts, and cryptojacking [3]. This is the area where Safe Chain integrates: running on install, not just at scan time.

The threat intelligence framing is accurate but narrow. Aikido’s own blog [5] positions this in the same category as Recorded Future, CrowdStrike, Mandiant, and Palo Alto Unit 42 — enterprise security products that cost tens of thousands a year per seat. That framing is useful for understanding what Aikido Intel is trying to be, but the comparison is a bit stretched. Those platforms do geopolitical threat tracking, actor attribution, and dark web monitoring. Aikido Intel does open-source supply chain intelligence. Both are “threat intelligence,” but the scope and audience are very different [5].

---

Features

Based on the homepage and published blog material:

Vulnerability feed:

• Monitors 4.3M+ packages across npm, PyPI, PHP, Ruby, NuGet, Maven, Rust, Go, C++, Dart, Elixir, Swift [homepage]
• LLM-powered analysis of changelogs and release notes to detect silently patched security issues [1]
• Cross-referenced against five vulnerability databases before assignment [1]
• Human security engineer review and severity assignment for each finding [1]
• Filterable by ecosystem, severity, and date at intel.aikido.dev [homepage]
• Assigns Aikido Vulnerability Numbers (AV-XXXX) for unfiled issues [1]
• CVE tracking: weekly re-evaluation to backfill CVE numbers as they’re assigned [1]

Malware feed:

• 125,200+ malicious packages catalogued [homepage]
• Covers npm ecosystem specifically for typosquatting, backdoors, account-hijack publishes [3]
• Package Health search UI at intel.aikido.dev/packages for on-demand lookups [homepage]

Safe Chain:

• Open-source npm install-time malware blocker [homepage]
• Checks packages against the Aikido Intel malware feed before execution
• Listed as a separate install: npm install safe-chain [homepage]

AGPL feed access:

• Vulnerability and malware data available under AGPL for direct use, modification, distribution [homepage]
• Developers can contribute findings back [homepage]

Commercial API:

• Designed for security vendors who want to embed the intelligence into their own products [homepage]
• Pricing not listed publicly — requires contacting sales [homepage]

What it does not do:

• It does not scan your codebase (that’s Aikido Security, the broader platform)
• It does not generate remediation advice or pull requests
• It does not integrate with CI/CD pipelines out of the box — you’d consume the API to build that [3]

---

Pricing: SaaS vs self-hosted math

This is where the picture gets unusual compared to typical self-hosted software reviews.

Aikido Intel itself is free. The web interface at intel.aikido.dev, the AGPL data feed, and Safe Chain are all available at no cost. You can browse vulnerabilities, search packages, and install Safe Chain without paying anything [homepage].

Commercial API access is a paid tier for companies that want to integrate the intelligence database into their own security products. Pricing is not publicly listed — the CTA says “Get Access” and routes to a contact form [homepage]. No ballpark figures are available in any of the reviewed sources.

Aikido Security (the broader platform) — which uses Intel as one of its data sources — has published pricing: free tier available, with paid plans running up to $1,050/month as reported by Tooliverse [3]. That’s the full SAST + SCA + DAST + runtime protection stack. For comparison: Snyk, one of the main competitors in the SCA/supply chain space, has paid plans starting around $25/developer/month. For a 10-developer team, that’s $250–300/month at entry level, scaling fast with usage.

Self-hosted reality: The AGPL license means you can technically fork and run the feed yourself, but the actual intelligence value comes from Aikido’s AI infrastructure processing 4.3M packages and their research team doing manual CVE triage. Self-hosting the AGPL codebase doesn’t give you their data pipeline. It gives you the code that could run a similar pipeline if you built one. For most teams, this distinction matters [homepage].

If you’re evaluating Aikido Intel purely as a free vulnerability lookup tool against paid alternatives like Snyk Advisor, OSS Index Pro, or a commercial threat feed, the cost math is simple: the feed is free, which is a strong argument. If you’re evaluating the commercial API against embedding an alternative feed in your product, pricing data is not available to make that comparison.

---

Deployment reality check

Aikido Intel is not a self-hosted application you stand up on a VPS. This is important to understand before evaluating it as “self-hosted software.”

What “self-hosted” means here:

• The AGPL license allows you to fork the GitHub repository (179 stars at time of writing) and run the data processing yourself [GitHub][homepage]
• In practice, the repository contains the feed infrastructure, not a turnkey Docker Compose you spin up in an afternoon
• The value proposition is the intelligence database and the AI/human research pipeline that populates it — running the code yourself without that data produces nothing useful

Safe Chain deployment is the only piece with a conventional installation path: npm install safe-chain added to your workflow, and it wraps your npm installs with malware checks. That’s a legitimate one-command deploy [homepage].

API integration for the commercial tier requires a proper REST integration — standard webhook/HTTP work for any developer, but it’s not drag-and-drop.

Potential friction points:

• The GitHub repo has no README text in the scraped data, which is unusual for an AGPL project inviting contributions [GitHub] — worth checking the actual repository before assuming self-hosting is documented
• With 179 GitHub stars, community troubleshooting resources are limited compared to projects like OSV Scanner or Grype
• Aikido Security’s broader platform requires manual tagging during initial setup for complex microservice architectures, per Tooliverse’s review of 42+ users who flagged this [3]

---

Pros and Cons

Pros

• Finds what CVE databases don’t. 67% of discovered vulnerabilities never filed anywhere — this is the core differentiator, and the stat is backed by a year of data from Aikido’s own feed [1].
• Free to use. The intelligence feed and web interface cost nothing. No signup required to browse intel.aikido.dev [homepage].
• 12+ ecosystems covered. npm, PyPI, Maven, NuGet, Go, Rust, and more — broad enough to cover most modern stacks [homepage].
• Malware detection is pre-install. Safe Chain runs at npm install time, not just at scan time — catches the package before it executes [homepage].
• AGPL license. More permissive than nothing, though less permissive than MIT. You can embed the code and data in your own tools with compliance obligations [homepage].
• Research-backed, not just automated. Human security engineers review findings before publication. The Axios example [1] demonstrates this catches real, high-impact issues in widely-used packages.
• Company behind it is established. Aikido Security has SOC 2 Type II and ISO 27001 compliance, 50,000+ customers for the broader platform [3]. Intel is a product from a real company, not an abandoned side project.

Cons

• Low GitHub star count (179). For an open-source security tool positioning itself as critical infrastructure, this is a weak adoption signal. Compare to Grype (8,000+ stars) or OSV Scanner (6,000+) [GitHub].
• AGPL, not MIT. If you’re building a commercial product on top of this, AGPL requires you to open-source your derivative. That rules out embedding it in proprietary products without a commercial license [homepage].
• Commercial API pricing opaque. No public pricing for the tier that most commercial teams would actually need. Requires a sales conversation [homepage].
• It’s a feed, not a scanner. You don’t point Aikido Intel at your repo and get a report. You consume the feed and integrate it into your existing tooling — or buy the broader Aikido Security platform [1][3].
• Repository quality unclear. No README available in the scrape data [GitHub]. An AGPL project with thin documentation creates barriers to the “freely use and modify” promise.
• Self-hosting the feed doesn’t get you the data. The intelligence value comes from Aikido’s processing pipeline and research team. You can run the code but you can’t run the data [homepage].
• Narrow scope. Aikido Intel covers open-source supply chain threats. It does not cover your first-party code (SAST), your cloud configuration (CSPM), or your runtime environment [3]. For most teams, this is one layer of defense, not a complete stack.
• Review coverage is limited. Unlike mature tools with hundreds of independent reviews, most available coverage is from Aikido’s own blog or the broader Aikido Security platform. Independent third-party reviews specific to Intel are sparse [3][4].

---

Who should use this / who shouldn’t

Use Aikido Intel if:

• You’re a developer or small security team that wants earlier warning on supply chain vulnerabilities than NVD/OSV provides, at zero cost.
• You’re building a security product and want to integrate a vulnerability and malware data source — and you’re willing to have a sales conversation about API pricing.
• You’re running Node.js-heavy workloads and want malware-on-install protection via Safe Chain.
• You already use or are evaluating Aikido Security as your AppSec platform — Intel is the intelligence layer that feeds it.

Don’t use it as a standalone solution if:

• You need a full vulnerability scanner that checks your code and dependencies against the feed. You need Aikido Security or a tool like Snyk, Grype, or Trivy for that workflow.
• You’re looking for a self-hosted drop-in you can deploy on a VPS in an afternoon. This is not that product.
• You’re evaluating strictly on community health — 179 stars is thin for a project you’d embed in production security infrastructure.

Pair it with something else if:

• You need SAST, secrets detection, or container scanning. Aikido Intel covers the open-source supply chain layer only [3].
• You want a CVE-based scanner with broader ecosystem tooling — use OSV Scanner or Grype for local scan workflows, and treat Intel as a supplementary early-warning layer.

---

Alternatives worth considering

For CVE/vulnerability feeds:

• OSV Database (Google) — free, open, covers most ecosystems, but only tracks disclosed CVEs. No silent-patch discovery [5].
• Snyk Advisor / Snyk Intel — similar commercial threat intelligence angle, but Snyk is a full-stack SCA tool and the intel feed is part of a broader paid platform. Known for mature SCA tooling and larger ecosystem coverage [4].
• GitHub Advisory Database — free, part of the GitHub Security Advisory ecosystem, powers Dependabot. Good coverage, but again only tracks filed CVEs [5].

For supply chain malware specifically:

• Socket Security — focuses specifically on malicious npm/PyPI packages, similar scope to the malware side of Aikido Intel, with a paid API tier and a more established open-source community presence.
• Phylum — open-source supply chain security feed with commercial API, comparable positioning.

For the broader AppSec stack:

• Aikido Security (full platform) — if you want Intel’s data plus SAST, DAST, container scanning, and IaC in one dashboard, this is the natural upgrade path [3].
• Snyk — the most widely deployed commercial SCA/supply chain tool. More expensive, more integrations, more mature at enterprise scale [4].
• Grype / Syft (Anchore) — open-source, MIT-licensed, strong for container and SBOM workflows. Narrower than Intel but easier to self-host meaningfully [4].

---

Bottom line

Aikido Intel fills a real gap: the 67% of security vulnerabilities that get patched without anyone telling you. For developers who’ve ever been burned by a dependency that was quietly fixed months before they upgraded, the feed is genuinely useful and free. Safe Chain adds a practical malware-on-install layer that takes about sixty seconds to set up in a Node.js project.

The limits are equally real. It’s a data feed, not a scanner. The “self-hosted” angle applies to the AGPL license but not to a deployable application. The GitHub star count is low for a security-critical tool, and commercial API pricing requires a sales call. For most non-technical founders, Aikido Intel alone won’t move the needle — you’d need the full Aikido Security platform or a comparable AppSec stack to act on what the feed tells you. What it does do well, it does better than the free alternatives. That’s a narrower value proposition than the homepage implies, but it’s a real one.

---

Sources

1. Mackenzie Jackson, Aikido Security Blog — “Meet Intel: Aikido’s Open Source threat feed powered by LLMs” (Dec 13, 2024). https://www.aikido.dev/blog/meet-intel-aikidos-open-source-threat-feed-powered-by-llms
2. Aikido Security — “How Aikido compares to GitLab Ultimate” (comparison page). https://www.aikido.dev/comparison/gitlab-ultimate-alternative
3. Tooliverse Editorial — “Aikido Security Review 2026 — AppSec Platform” (Verified Mar 16, 2026). https://tooliverse.ai/tools/aikido-security
4. OX Security Blog — “Top 5 Aikido Alternatives for Application Security Management (2026)” (Dec 17, 2025). https://www.ox.security/blog/aikido-alternatives/
5. Divine Odazie, Aikido Security Blog — “The Top 7 Threat Intelligence Tools in 2026” (Dec 15, 2025). https://www.aikido.dev/blog/the-top-7-threat-intelligence-tools

Primary sources:

• Aikido Intel homepage: https://intel.aikido.dev
• GitHub repository: https://github.com/aikidosec/intel (179 stars, AGPL license)
• Package health search: https://intel.aikido.dev/packages