Datadog Security Labs analyzes leaked TeamPCP "Shai-Hulud" offensive framework source code

UPDATE (2026-05-13 — follows TeamPCP coverage 2026-05-13): Datadog Security Labs published an analysis of the TeamPCP "Shai-Hulud" offensive worm source code on 2026-05-13, after the complete framework was briefly accessible as a public GitHub repository on 2026-05-12 before the account was removed (Datadog Security Labs, 2026-05-13). The brief public exposure gave researchers direct visibility into the worm's internal architecture: it is a TypeScript/Bun toolkit that automates GitHub Actions pwn-request exploitation — specifically targeting pull_request_target workflows that perform unsanitized checkouts — to harvest OIDC tokens and GITHUB_TOKEN values, then propagate across npm packages using the stolen credentials. The automation is fully self-contained; victim-repository selection is not manually guided, consistent with the worm-class spread observed in the original TanStack campaign. The leaked code also exposes the environment-variable injection technique (${{ github.event.pull_request.head.sha }} substitution in run steps) as a key primitive. Defenders should not execute the leaked code. The architectural disclosure accelerates defensive posture: prioritise auditing pull_request_target triggers with checkout steps in the same job, review OIDC token permission scopes, and apply environment variable sanitization. MITRE ATT&CK: T1195.002 (Compromise Software Supply Chain), T1552.001 (Credentials in Files), T1059.004 (Unix Shell).