/
← Voltar ao feed
high technology cloud cryptocurrency cloud services cloud_services artificial_intelligence

Railway PaaS breach exposes customer environment variables and secrets; deployments paused, wallets drained, all users told to rotate keys

3 fonte(s) · 3 atualização(ões) · publicado 22/10/2026 08:15 · coletado há 6 horas

Matéria consolidada. Reúne 3 matérias que o Tifon tinha publicado separadamente sobre a mesma história. A IA reescreveu o relatório com o que as 3 fontes trouxeram em 23/10/2026 00:05; os links antigos trazem para cá.

Fontes: bleepingcomputer.com ↗ techcrunch.com ↗ thehackernews.com ↗

Railway, a Platform-as-a-Service provider, disclosed "anomalous activity" on the systems that store customer environment variables. An unauthorized third party may have accessed encrypted variables and, in some cases, obtained the decrypted contents (thehackernews.com, techcrunch.com, bleepingcomputer.com). Railway detected the activity earlier in the week of 22 October 2026 (bleepingcomputer.com). Initial analysis points to a "subset of users", but Railway urged all users to rotate secrets and paused customer deployments, which effectively froze customer sites and apps (techcrunch.com, bleepingcomputer.com). Railway says the incident is likely part of a series of attacks on platform companies aimed at compromising AI inference providers, and it notified those providers so they could revoke keys. Security researcher Taylor Monahan reported that cryptocurrency wallets whose private keys were stored on Railway were drained (thehackernews.com). The root cause, initial access vector, attacker identity and number of affected accounts have not been disclosed. Railway says it will notify users directly once their accounts are confirmed compromised (bleepingcomputer.com).

Detalhes técnicos

  • Exposed data: environment variables that typically hold API keys, database credentials and private registry credentials. Exposed secrets could let an attacker authenticate and access data as the legitimate owner (thehackernews.com, techcrunch.com, bleepingcomputer.com).
  • Scale of the platform: about 4.8 million users and more than 16 million hosted services (bleepingcomputer.com).
  • Wallet theft: Monahan reported the first known theft at 7:36 p.m., at least two hours before official disclosure, and an attacker-linked address 0xd9773325282874d18cb83f1fe50f14c39750b2ee (thehackernews.com).
  • Railway's response: it rotated all Railway bucket credentials and invalidated Railway API tokens (thehackernews.com). Builds were restored for Pro and Enterprise workspaces, but deployments were delayed with many queued. Hobby, Free and Trial workspaces were still being restored, with some builds failing (thehackernews.com, bleepingcomputer.com).
  • Related incident: OpenCode, an open source AI coding tool, reported a similar incident earlier in the week. More than 1% of its customers' API keys and secrets were taken, and OpenCode attributed the attack to AI distillation (techcrunch.com).
  • Context, not linked to Railway: CISA recently warned of large-scale attacks by China-based AI companies against U.S. firms to improve their own models (techcrunch.com). The incident also appears separate from the PHP2Shell PHP vulnerabilities, which were patched the same day; Amazon researchers reported exploitation evidence four days before the patches (techcrunch.com).
  • Related LLMjacking research, reported in the same coverage and not shown to be tied to Railway (thehackernews.com):
    • Wiz documented sponsored Google Search ads for terms like "google cloud" or "gap" that redirected users via Google Sites to a fake Google login page harvesting credentials and MFA codes. Attackers then abused excessive privileges, linked attacker-controlled external GCP projects to the victim's billing accounts, added attacker Gmail accounts to the billing account, and ran AI workloads in foreign projects, hiding prompts and outputs from the victim.
    • The Symantec Threat Hunter Team reported an offensive AI agent in "Skynet mode" that targets exposed LLM endpoints. It scans the internet, exploits vulnerabilities, steals credentials, harvests secrets and mines cryptocurrency autonomously for up to 100,000 steps. Targeted deployments include Microsoft Visual Studio Code, Django, Jupyter, Laravel, Supabase, Next.js and Flask/Werkzeug.
  • No attribution, IOCs beyond the wallet address above, or ransom demand have been reported (techcrunch.com).

Remediação

  • Revoke and rotate all environment variables and secrets stored on Railway, including API keys, cloud credentials, database credentials and private registry credentials, even if you have not been notified of compromise (all sources).
  • Review service logs and downstream services (cloud storage, databases, third-party APIs) for unauthorized access using those credentials.
  • Rotate any cryptocurrency wallet private keys that were stored on Railway (thehackernews.com).
  • AI inference providers should revoke all their keys, as Railway advised (bleepingcomputer.com).
  • Follow Railway's blog post and status updates for restoration of deployments (techcrunch.com).

Relatos divergentes

  • Scope: Railway's initial analysis points to a "subset of users" (thehackernews.com, techcrunch.com, bleepingcomputer.com), yet it asks all users to rotate secrets. No source gives a number of affected users, so the real scope remains unconfirmed.
  • Notification: thehackernews.com says Railway notified upstream providers to revoke exposed keys proactively. bleepingcomputer.com, the most recent source, says AI inference providers were notified directly and told to revoke all their keys.
  • Motive: Railway links the breach to a campaign against AI inference providers. The theft of wallet funds (thehackernews.com) shows exposed secrets were also abused for financial gain. OpenCode's attribution to AI distillation (techcrunch.com) has not been connected to Railway by any source.
  • Restoration: bleepingcomputer.com gives the state as of 22 October 08:44 UTC. Builds for Pro and Enterprise were restored, but Hobby, Free and Trial had not resumed. thehackernews.com describes those lower tiers as still being restored. The sources are consistent, and no later update is available.

Linha do tempo

  1. bleepingcomputer.com ↗ · pub. 22/10/2026 18:25 · coletado 22/10 20:00

    Railway warns users to rotate secrets after possible theft of decrypted environment variables

    ler a análise desta fonte

    Cloud hosting provider Railway warned users that an unauthorized third party may have accessed encrypted environment variables and, in some cases, obtained their decrypted contents. Railway says the activity was part of a series of attacks targeting AI inference providers, which were notified directly and told to revoke all their keys. The platform has about 4.8 million users and hosts more than 16 million services. Railway urges all users to rotate secrets immediately, even though initial analysis suggests only a subset is affected. It also paused all deployments while the investigation continues.

    Technical Details

    • Environment variables on Railway can hold API keys, database credentials and private registry credentials.
    • Railway detected suspicious activity on the systems that store customer environment variables earlier in the week. It has not disclosed the initial access vector, the attacker, or the number of affected accounts.
    • As of October 22 at 08:44 UTC, builds were restored for Pro and Enterprise workspaces, but deployments were delayed with many queued. Builds for Hobby, Free and Trial workspaces had not yet resumed.
    • Railway will notify users directly once their accounts are confirmed compromised.

    Remediation

    • Rotate all environment variables and secrets stored on Railway, including private registry credentials.
    • Check service logs for unusual activity.
    • AI inference providers should revoke all their keys, as Railway advised.
  2. techcrunch.com ↗ · pub. 22/10/2026 11:03 · coletado 22/10 13:00

    Railway confirms breach exposing customer environment variables, urges all users to rotate keys and pauses deployments

    ler a análise desta fonte

    Railway, a cloud development platform, confirmed that unidentified attackers may have accessed encrypted environment variables and, for some, obtained the decrypted contents. These variables typically hold passwords and API keys that apps use to reach cloud storage and databases. Railway's initial analysis points to a "subset of users", but it urged all users to revoke and rotate their keys, and it paused customer deployments during the investigation, which effectively froze customer sites and apps. The number of affected users, the root cause and the compromised systems have not been disclosed.

    Technical Details

    • Railway said the breach was likely part of a broader campaign targeting platforms "with the aim of compromising AI inference providers".
    • Exposed customer secrets could let an attacker authenticate and access data as the legitimate owner.
    • TechCrunch notes that OpenCode, an open source AI coding tool, reported a similar incident earlier in the week. More than 1% of its customers' API keys and secrets were taken, and OpenCode attributed the attack to AI distillation.
    • TechCrunch also reports that CISA recently warned of large-scale attacks by China-based AI companies against U.S. firms to improve their own models. The article does not link this warning to Railway.
    • The incident appears separate from the PHP2Shell PHP vulnerabilities, for which patches were released the same day. Amazon researchers reported exploitation evidence four days before the patches.
    • No attribution, IOCs or ransom demand have been reported.

    Remediation

    • Revoke and rotate all API keys, cloud credentials and other secrets stored as Railway environment variables.
    • Review downstream services (cloud storage, databases, third-party APIs) for unauthorized access using those credentials.
    • Follow Railway's blog post and status updates for the restoration of deployments.
  3. thehackernews.com ↗ · pub. 22/10/2026 08:15 · coletado 22/10 09:00 primeiro registro

    Railway PaaS breach exposes customer environment variables and secrets; crypto wallets drained, Wiz and Symantec report LLMjacking

    ler a análise desta fonte

    Railway, a Platform-as-a-Service provider, disclosed "anomalous activity" affecting systems that store customer environment variables. An unauthorized third party may have accessed encrypted variables and in some cases obtained the decrypted contents, affecting a "subset of users" (number undisclosed). Railway said the incident is likely part of a series of attacks on platform companies, with the ultimate goal of compromising AI inference providers, and has notified upstream providers so they can proactively revoke exposed keys. Security researcher Taylor Monahan reported that attackers drained cryptocurrency wallets whose private keys were stored on Railway, with the first known theft at 7:36 p.m., at least two hours before official disclosure. The article also covers related LLMjacking activity reported by Wiz and the Symantec Threat Hunter Team.

    Technical Details

    • Exposed data: environment variables such as API keys, database credentials and private registry credentials. Attacker-linked address reported by Monahan: 0xd9773325282874d18cb83f1fe50f14c39750b2ee.
    • Railway rotated all Railway bucket credentials and invalidated Railway API tokens. Builds were restored for Pro and Enterprise workspaces; Hobby, Free and Trial are still being restored, with some builds failing and deployments queued.
    • Wiz documented a related LLMjacking case: sponsored Google Search ads for terms like "google cloud" or "gap" redirected users via Google Sites to a fake Google login page that harvested credentials and MFA codes. Attackers then abused excessive privileges, linked external attacker-controlled GCP projects to the victim's billing accounts and added attacker Gmail accounts to the billing account. AI workloads ran in foreign projects, hiding prompts and outputs from the victim.
    • Symantec reported an offensive AI agent operating in "Skynet mode" that targets exposed LLM endpoints to scan the internet, exploit vulnerabilities, steal credentials, harvest secrets and mine cryptocurrency autonomously for up to 100,000 steps. Targeted deployments include Microsoft Visual Studio Code, Django, Jupyter, Laravel, Supabase, Next.js and Flask/Werkzeug.

    Remediation

    • Rotate all environment variables stored in Railway, including API keys, and private registry credentials.
    • Review logs for affected services for unusual activity.
    • Rotate any cryptocurrency wallet private keys that were stored on Railway.