I wrote a custom jellyfin addon to get back access to ssh

Full disclosure: I work on community at Always Further, the team behind this. Not the author. Posting because Luke's approach to tackling this challenge is unique and of an interest to the netsec community.

The core idea: if an AI agent is compromised, any log the agent itself writes becomes part of the attack surface. The post walks through how they split auditing into a supervisor process the sandboxed child can't reach, then uses the same Merkle tree + hash-chain construction RFC 6962 (Certificate Transparency) uses to make edits, truncation, and reordering all detectable.

There's a concrete threat-model table near the end that lists what each attack looks like and what structurally stops it. Worth skipping to if you don't want the crypto primer.

Bitwarden CLI npm package got compromised today, looks like part of the ongoing Checkmarx supply chain attack

If you’re using @bitwarden/cli version 2026.4.0, you might want to check your setup

From what researchers found:

- malicious file added (bw1.js)

- steals creds from GitHub, npm, AWS, Azure, GCP, SSH, env vars

- can read GitHub Actions runner memory

- exfiltrates data and even tries to spread via npm + workflows

- adds persistence through bash/zsh profiles

Some weird indicators:

- calls to audit.checkmarx.cx

- temp file like /tmp/tmp.987654321.lock

- random public repos with dune-style names (atreides, fremen etc.)

- commits with “LongLiveTheResistanceAgainstMachines”

Important part, this is only the npm CLI package right now, not the extensions or main apps

If you used it recently:

probably safest to rotate your tokens and check your CI logs and repos

Source is Socket research (posted a few hours ago)

Curious if anyone here actually got hit or noticed anything weird