tag: Developer-Tools · 11 items
- Engineer — Act: A trojanized Claude Code extension targeting Cursor/VS Code installs a persistent SSH backdoor on developer machines — audit all installed IDE extensions for this package and follow Microsoft’s remediation steps to remove backdoor artifacts and rotate any credentials accessible from affected workstations.
- SOC/IR — Act: Microsoft’s analysis includes detection guidance; sweep developer endpoints for the malicious extension presence and hunt for anomalous outbound SSH connections or new authorized_keys entries on workstations since late October 2026.
- Leader — Act: A supply-chain compromise of developer tooling with persistent backdoor access is a same-week risk event — confirm whether engineers use Cursor/VS Code extensions from untrusted sources, assess any exposure, and consider a brief to engineering leadership on extension vetting policy.
- Engineer — Act: If your team uses the Claude Code extension on Cursor, audit all affected machines for version 2.1.313, check ~/.ssh/authorized_keys and running SSH tunnel processes for unauthorized entries, and rotate any credentials accessible from those developer workstations.
- SOC/IR — Act: Hunt across developer endpoints for the malicious extension version 2.1.313 and sweep for unauthorized SSH key additions or anomalous outbound SSH sessions from developer machines since the update was distributed; treat any hit as a potential persistent-access incident requiring IR.
- Leader — Act: Determine whether any engineers in your organization use the Claude Code extension on Cursor; if so, initiate an IR review — a supply-chain compromise on developer tooling means potential persistent access to source code, secrets, and cloud credentials, which warrants a leadership brief before it surfaces externally.
- Engineer — Act: Confirmed supply-chain compromise of a developer CLI tool; audit all developer machines for the backdoored OVHcloud CLI version and immediately rotate any API keys, tokens, or credentials that the tool could have accessed.
- SOC/IR — Plan: Active info-stealer deployed via compromised CLI; without published IOCs, queue a hunt for anomalous credential access or exfiltration activity originating from developer endpoints that run cloud CLI tooling.
- Leader — Act: A cloud provider’s official CLI was weaponized to steal developer secrets — confirm this week whether engineering teams use OVHcloud CLI and request OVHcloud’s incident report detailing which versions were affected and what credentials may be at risk.
- Engineer — Act: A public PoC is available for this sandbox escape in Claude Code; any team running Claude Code in dev workflows or CI should update to the patched release listed in GHSA-vp62-r36r-9xqp immediately.
- SOC/IR — Skip
- Leader — Skip
- Signals: CVE-2026-39861 — CISA KEV: not listed, EPSS 0.01, public PoC on GitHub
- Engineer — Learn: A local reverse-proxy that redacts sensitive tokens before requests reach cloud LLMs (Cursor, Claude Code, Codex) and restores them in the response stream — worth evaluating if your team pipes proprietary code or credentials through AI coding assistants, but no exploitation or urgency signals exist.
- SOC/IR — Skip
- Leader — Learn: Highlights a real risk vector — developers sending sensitive code and secrets to third-party LLM providers — but this is a single trending OSS tool, not an incident; useful framing for a future AI-tool data-handling policy conversation, not same-week action.
- Engineer — Plan: If your team uses DeepSeek Harness for AI coding agents, verify you are on a patched version and audit whether agents have been granted excessive tool permissions; no active exploitation signals yet, but the bypass requires only a single command.
- SOC/IR — Learn: Illustrates a novel attack class where an AI agent escapes its sandbox via the harness’s own API — no IOCs or active exploitation to hunt for today, but worth tracking as AI coding agent adoption grows.
- Leader — Learn: Concrete example of AI agent trust boundary failure useful for AI security governance discussions; no board-level event, but reinforces the need for policy on approved AI coding tools before they proliferate.
- Engineer — Learn: If you expose Solana JSON-RPC or gRPC dev endpoints (e.g., surfpool) on public interfaces, audit firewall rules to ensure they are not internet-reachable; no active exploitation or PoC reported.
- SOC/IR — Learn: Awareness item: opportunistic scans targeting Solana dev endpoints are occurring, but no IOCs, TTPs, or confirmed exploitation are provided to act on.
- Leader — Skip
- Engineer — Act: If your team uses Open VSX-sourced extensions (common in Theia, Gitpod, or VS Code OSS environments), audit installed extensions against the 77 removed packages and remove any installed between July 26–August 1, 2026; check build/dev environments for unexpected outbound connections during that window.
- SOC/IR — Act: Hunt for anomalous outbound traffic from developer workstations and CI runners between July 26 and August 1, 2026 that may indicate data exfiltration from compromised extensions; correlate against Open VSX extension install events in endpoint logs.
- Leader — Plan: This incident illustrates supply-chain risk in developer tooling marketplaces; work with engineering leads this quarter to establish an approved-extension policy and inventory for IDE plugins used across the org.
- Engineer — Act: If your team uses Open VSX (common in VS Code OSS or VSCodium environments), audit installed extensions against the removed list and purge any matches; review extension installation policies in CI/CD or dev container configs to restrict to known-good sources.
- SOC/IR — Plan: Build or tune detections for unexpected outbound connections from IDE processes (code, codium) to unknown endpoints; consider hunting for extension-related network activity in EDR telemetry from developer workstations over the past 90 days.
- Leader — Learn: This incident illustrates ongoing supply-chain risk in developer tooling marketplaces; useful context for evaluating software vetting policies in engineering onboarding, but no immediate leadership action required.
- Engineer — Learn: If Codex is in your development toolchain or CI pipelines, review the repository for security boundaries, sandboxing limitations, and trust assumptions — no exploit pressure, but 536 HN upvotes suggests substantive security guidance worth absorbing.
- SOC/IR — Skip
- Leader — Learn: If developers in your organization use OpenAI Codex, this repository likely clarifies the product’s security posture and responsible-use boundaries — useful context for an AI tool risk policy, but no immediate action required.
- Engineer — Act: Public PoC exists for a symlink-based sandbox escape in Claude Code, which engineers and CI/CD pipelines commonly run; update Claude Code to the patched release immediately and audit any pipelines that invoke it with elevated filesystem access.
- SOC/IR — Learn: Low EPSS (0.01) and no active exploitation campaign; no IOCs or ATT&CK-mappable TTPs are provided, but the symlink sandbox-escape technique is worth noting if Claude Code runs in monitored developer environments.
- Leader — Skip
- Signals: CVE-2026-39861 — CISA KEV: not listed, EPSS 0.01, public PoC on GitHub