Mandiant Warns Developers of Supply Chain Attacks on IDEs and CI/CD

Mandiant has published guidance on hardening software development environments amid recent supply-chain campaigns targeting trusted security scanners, utility libraries and AI developer tools. The company says attackers are also targeting developer workstations and IDEs through social engineering, malicious extensions and typosquatted dependencies to steal personal access tokens, SSH keys, API tokens and session credentials.

More advanced techniques include GitHub Actions cache poisoning, OpenID Connect (OIDC) token extraction and changing mutable action tags to distribute compromised packages that retain legitimate cryptographic provenance. The guidance describes a defence-in-depth model covering developer endpoints, repositories, artefact management, CI/CD and deployment.

Recommended measures include using approved and version-pinned IDEs and extensions, containerised or virtualised development environments, local secret scanning, phishing-resistant MFA and conditional access. Mandiant advises replacing long-lived credentials with short-lived, narrowly scoped tokens, using GitHub Apps for automated integrations, pinning dependencies and container images to exact versions or SHA-256 digests, and enforcing signed provenance and SBOM attestations.

It also recommends a seven-day release-age cooldown for newly published public packages, central proxies and quarantine controls, continuous vulnerability scanning, and VEX statements to reduce irrelevant alerts.

For CI/CD, the guidance calls for ephemeral single-use runners, network restrictions, manual approval for untrusted pull-request code, and cache isolation to prevent poisoned dependencies reaching production builds. OIDC-based federated identities, least-privilege permissions, immutable branch histories and early scanning gates are also advised.

At deployment, signed-image admission controls, hardened workloads, just-in-time access, microsegmentation, continuous posture monitoring and detailed logging should help contain attacks that bypass earlier safeguards.