Jordan: Someone leaked a commercial-grade iPhone exploit kit on GitHub this weekend. Publicly. For free. If you have executives still running iOS 16, that's not a theoretical problem anymore — that's an active exposure. We'll get to that. But first, we need to talk about self-propagating malware poisoning open source packages and wiping machines. Today is a heavy one.

Alex: This is Cleartext. Tuesday, March 24th, 2026. I'm Alex Chen.

Jordan: And I'm Jordan Reeves. Here's what we're covering today: a geopolitically targeted supply chain wiper that's inside development environments right now. A major supply chain attack on Trivy — yes, a security tool — spreading through Docker and GitHub. AI-powered phishing that's already compromised hundreds of organizations. An alleged Lockheed Martin breach with an adversarial extortion twist. Personal CISO liability getting worse before it gets better. And that iPhone exploit kit. Let's move.

Alex: We're leading with the open source wiper story because the combination of factors here is unlike anything we've seen in a while. This malware is self-propagating — it spreads through poisoned open source packages — and it carries a geopolitical targeting mechanism. Machines it detects as Iran-based get wiped. That's not ransomware. That's a weapon with a specific political address on it.

Jordan: The geolocation-based payload is what makes this notable from an intelligence standpoint. Someone built a kill condition into the code. That's either a nation-state actor or someone very deliberately trying to look like one. Either way, the collateral damage risk is real. If your developers are pulling infected packages into their environments, the wiper logic fires based on IP-detected location — but network routing is messy, VPNs exist, and attribution is imperfect. You could have machines in your own environment that trigger unexpected behavior.

Alex: The CISO action here is immediate and it's not complicated: audit your dependency trees. Check your software composition analysis tools for any packages that have been flagged or updated in the last week without a clear provenance trail. Development environments are notoriously under-monitored from a security perspective — developers need speed, and security friction gets bypassed. This is why that shortcut has a price.

Jordan: And this connects directly to story four, so let's go there. Because while you're thinking about your supply chain hygiene, the Trivy attack is a punch in the stomach for anyone who thought their security tooling was a safe harbor.

Alex: Trivy is Aqua Security's open source vulnerability scanner. Widely used in CI/CD pipelines. Hundreds of thousands of deployments. The TeamPCP group compromised it, pushed malicious Docker images, and hijacked Aqua's GitHub organization to tamper with dozens of repositories. What they were after: CI/CD secrets. Cloud credentials. SSH keys. API tokens. The crown jewels of your automated infrastructure.

Jordan: This is the move that sophisticated actors are making now. They're not kicking in the front door. They're compromising the tools that your security team trusts implicitly. When your scanner is the vector, your detection capability is what's been turned against you. That's not a vulnerability — that's a strategic inversion.

Alex: If you're running Trivy in your pipeline, verify the integrity of your installation against known-good hashes, check your GitHub Actions for tampered workflows, and rotate any secrets that pipeline has touched. Assume breach posture on this one until you've confirmed clean.

Jordan: Let's talk about the AI phishing story because the numbers from Huntress are striking. Hundreds of organizations confirmed compromised, and they're explicitly saying that's probably a fraction of total victims. What's different here isn't just that AI is being used to write better lures — it's scale. The economics of high-quality phishing have collapsed.

Alex: From a board conversation standpoint, this changes the way you have to frame email security investment. For years we've told employees to look for awkward phrasing, grammatical errors, generic greetings. That heuristic is dead. AI-generated phishing is grammatically perfect, contextually aware, and can be personalized at industrial scale. Your email gateway and your awareness training both need to be reassessed against this threat model, not last year's.

Jordan: The detection emphasis has to shift upstream — toward behavioral analytics, toward link and payload analysis, toward identity anomaly detection post-click — because content quality alone no longer discriminates malicious from legitimate.

Alex: Now, the Lockheed Martin allegation. A pro-Iran hacktivist group is claiming they've breached Lockheed, and the demand is notable: pay up, or we sell the data to U.S. adversaries. That framing is significant.

Jordan: It blurs the line between hacktivism and espionage-for-hire in a way that's been building for a few years. Iran's cyber ecosystem has increasingly operated in this gray zone — ideologically motivated actors who are also willing to monetize access. The extortion-to-adversary-sale model is a direct threat to national security equities and it puts Lockheed's board in an impossible position. You can't negotiate with someone who's offering your data to your geopolitical adversaries.

Alex: For CISOs in the defense industrial base and the broader supply chain, the lesson is about your data classification and compartmentalization posture. What would a breach of your environment actually expose? If the answer is "I'm not sure," that's a gap that needs to close now. Lockheed can weather this reputationally. A Tier 2 supplier probably cannot.

Jordan: FBI also dropped warnings yesterday about Russian and Iranian campaigns targeting Signal and Telegram users — dissidents, journalists, opposition figures primarily. This isn't a direct enterprise threat for most organizations, but if you have employees in those communities, or if you're in media, NGOs, or international operations, your communications security policy should reflect this.

Alex: And this is a reminder that your personal devices policy matters. Executive communications on personal messaging apps is a persistent governance gap. The FBI warning is a nudge to close it.

Jordan: Quick note on the DOE five-year grid hardening plan. They've outlined a solid OT security and resilience roadmap. The problem, as analysts are pointing out, is the funding environment and workforce constraints make execution a real question. For energy sector CISOs and critical infrastructure operators, don't wait for federal partnership to deliver. That roadmap tells you where the expectations are heading — build toward it on your own timeline.

Alex: On governance — and this one is personal to a lot of our listeners — the CISO liability trend is accelerating. Regulators are pursuing individual accountability after major breaches, and it is visibly reshaping how CISOs report risk to boards. And not in a healthy way. When the fear of personal liability drives you to over-document defensively rather than communicate clearly, security culture weakens.

Jordan: The chilling effect on talent is real. Experienced practitioners are passing on CISO roles or demanding indemnification structures that boards aren't used to negotiating. If you're in that conversation, know what you're negotiating for: D&O coverage that explicitly covers regulatory actions, clearly documented board-level risk acceptance, and employment agreements that delineate your authority alongside your accountability.

Alex: Healthcare CISOs — brief but important. The HIPAA Security Rule overhaul may finalize as early as May 2026. This is the first major update in decades and it's being anchored to modern frameworks. If you haven't started your gap assessment against the proposed rule, the window is closing.

Jordan: On the iPhone exploit kit — DarkSword, now publicly available on GitHub. This targets older iOS versions and installs spyware. The barrier to deploy this just dropped to zero. The mitigation is straightforward: enforce iOS updates through your MDM. If you have executives, board members, or anyone with access to sensitive information running iOS below current, that's a priority ticket today, not next patch cycle.

Alex: Looking at this week as a whole, the theme is trust infrastructure under attack. Your open source dependencies, your security tools, your messaging platforms, your developer pipelines. Attackers have figured out that the highest-leverage entry points aren't the endpoints you're watching — they're the systems and processes you rely on to watch everything else.

Jordan: The strategic implication for CISOs going into Q2 is this: your supply chain security program needs to be resourced like your endpoint security program. Software composition analysis, CI/CD pipeline integrity, third-party tool vetting — these are not nice-to-haves. This week proved that.

Alex: And the geopolitical temperature is elevated. Three of today's nine stories have direct Iranian nexus. The overlap between hacktivism, state-sponsored activity, and criminal monetization in that ecosystem is getting tighter. Watch that space closely over the next several weeks.

Jordan: We will be.

Alex: That's Cleartext for Tuesday, March 24th. If something from today's episode is worth bringing to your board or your leadership team, we've done our job. We'll be back tomorrow. Stay sharp.