Loading story…
Anthropic’s Claude shared-chat pages were reportedly indexed by Google and Bing, making publicly shared conversations discoverable through searches such as site:claude.ai/share. The exposed material reportedly included legal discussions, engineering and software development work, resumes, business conversations, personal exchanges, and in some accounts credentials and cryptocurrency-related data. Reports said the pages were publicly accessible snapshots intended for anyone with the link, and that missing or ineffective noindex controls likely allowed search engines to crawl and list them once the URLs appeared on public websites.
The incident was described as a privacy exposure rather than a direct compromise of Anthropic’s systems, but it raised concerns over data leakage, intellectual property exposure, and compliance risk. By Sunday, many of the affected search results had reportedly disappeared, suggesting deindexing or a backend change, though previously saved share URLs could still remain reachable unless revoked server-side. Anthropic had not issued a public statement at the time of reporting, and users were urged to review shared chats, remove sensitive links, and assume any AI conversation shared publicly could become searchable on the open web.
Mallory correlates global threat intelligence with your attack surface — know if you’re exposed before adversaries strike.
2 events from the most recent confirmed update back to the earliest known activity.
By Sunday, many of the Google results for indexed Claude shared chats had reportedly been removed, suggesting deindexing or a backend change. Reports noted that previously saved share URLs could still remain accessible unless Anthropic revoked them server-side.
Publicly shared Claude conversation URLs were discoverable through search queries such as "site:claude.ai/share," exposing chats that included business discussions, software development work, personal information, and other sensitive content. Reporting said the pages were publicly accessible and appeared to lack noindex controls, allowing standard search crawlers to index them once links were exposed online.
Vulnerabilities, threat actors, malware, products, organizations, and breaches Mallory has linked to this story.
11 references tracked. Mallory keeps watching after this page renders.
malwarebytes.com
Open sourcehackread.com
Open sourcefortune.com
Open sourcereddit.com
Open sourcethecybersecguru.com
Open sourcecybersecuritynews.com
Open sourcecryptika.com
Open sourcereddit.com
Open sourceMap indicators from this story to your assets and identify affected systems in minutes.
Every observed campaign, victim, and pivot linked to actors named in this story.
Malware, exploits, and IOCs connected to the activity described here.
YARA, Sigma, and Snort rules deployed to your SIEM as soon as they’re published.
Get matching new stories delivered to your team as they break — not the next morning.
Ask questions about this story and take action on the answers.
Other stories Mallory is tracking that overlap on entities, sources, or topic labels.
Anthropic confirmed that an internal data exposure revealed details about **Claude Mythos**, an unreleased AI model the company described in leaked draft materials as its most capable system to date. The leak surfaced through an unsecured, publicly searchable cache tied to a content management system misconfiguration, which reportedly failed to mark uploaded files as private before storing them in a publicly accessible data lake. Exposed materials included a draft blog post, PDFs, images, release-related content, and information about an exclusive CEO-level event. Reports said the exposure involved nearly **3,000 unpublished assets** and included internal language stating that Anthropic had assessed Claude Mythos as posing **unprecedented cybersecurity risks**. Anthropic said the model is real and is being tested by early-access customers, but it had not disclosed whether anyone beyond journalists accessed the data or what remediation steps were taken. The incident has intensified scrutiny of the company’s data governance, access controls, and broader safety claims around pre-deployment evaluation of advanced AI systems.
Jun 29, 2026
Researchers at **Oasis Security** disclosed an attack chain dubbed **"Claudy Day"** that combines three flaws in Anthropic’s **Claude** platform to enable covert data theft. The chain starts with attacker-controlled Google search results or fake Claude ads that abuse an **open redirect** on `claude.com`, leading victims to what appears to be a legitimate Claude page. From there, a pre-filled chat URL can carry **hidden prompt-injection instructions** embedded in HTML so the user sees benign text while Claude processes malicious commands in the background. According to the reports, the chained flaws allow attackers to silently instruct Claude to search prior conversations for sensitive information and exfiltrate it through the **Anthropic Files API**, creating an end-to-end path from victim luring to unauthorized data exposure. The issue is notable because it does not require extra tooling, integrations, or MCP servers; built-in Claude features are sufficient for exploitation. A separate report about Claude autonomously exploiting vulnerable websites is **not** the same incident, and a report on the **SideWinder** espionage campaign is unrelated.
Jun 29, 2026
Anthropic accidentally exposed the source code of its Claude Code CLI after publishing npm package version `2.1.88` with a large JavaScript source map that pointed to unobfuscated TypeScript files hosted in company-controlled storage. Researcher **Chaofan Shou** identified the issue, which exposed roughly **1,900 to 2,000 files** and more than **512,000 lines of code** before the package was removed. Anthropic said the incident was caused by a manual release-packaging error rather than an external intrusion, and stated that no customer data or credentials were leaked. The code was quickly mirrored and forked on GitHub, making containment difficult, while Anthropic later acknowledged that automation gaps in its deployment process contributed to the exposure. Analysis of the leaked code revealed internal architecture and unreleased capabilities including **KAIROS** background autonomy, persistent memory features, multi-agent orchestration, hidden feature flags, undocumented commands, and controls intended to resist model distillation or disclosure of internal names. Security researchers warned that exposing Claude Code’s permission logic, trust boundaries, prompts, and execution pathways could help attackers craft jailbreaks, prompt-injection lures, malicious repositories, and persistence techniques against the tool. Anthropic’s response also drew scrutiny after a DMCA notice led GitHub to disable thousands of legitimate forks before the company asked for narrower enforcement, and the leak’s impact was compounded by reports of typosquatted npm packages and a separate compromised `axios` dependency affecting some npm installs during the same period.
Jun 29, 2026