diff --git a/src/pentesting-ci-cd/github-security/abusing-github-actions/README.md b/src/pentesting-ci-cd/github-security/abusing-github-actions/README.md
index 525db8c725..578ab43b2e 100644
--- a/src/pentesting-ci-cd/github-security/abusing-github-actions/README.md
+++ b/src/pentesting-ci-cd/github-security/abusing-github-actions/README.md
@@ -736,6 +736,10 @@ import base64,os;exec(base64.b64decode(os.environ["STAGE2_B64"]))
Drop the line above into a file such as `evil.pth` inside `site-packages` and it will execute during Python startup. This is especially useful in build agents that continuously spawn Python tooling (`pip`, linters, test runners, release scripts).[[25]](#references)
+A poisoned wheel can preserve normal package behavior while adding a launcher to its `__init__.py`. The launcher runs when the package is imported, not merely when the wheel is installed; the compromised Lightning wheel used a hidden `_runtime` directory for this path.[[49]](#references)
+
+A wheels-only policy excludes source builds, but it does not block import-time code or `.pth` startup hooks shipped in wheels.[[50]](#references)
+
#### npm supply-chain pivots from GitHub Actions
For `binding.gyp` / Phantom Gyp execution, wormable npm publishing with stolen CI identities, and the limits of trusted publishing provenance after workflow compromise, check:
@@ -997,4 +1001,6 @@ An organization in GitHub is very proactive in reporting accounts to GitHub. All
- [46] [actions/toolkit core package - GitHub](https://github.com/actions/toolkit/tree/main/packages/core#setting-a-secret)
- [47] [actions/checkout v5 action manifest](https://github.com/actions/checkout/blob/e8d4307400f9427dba7cb98e488d6ab85f1cec5f/action.yml#L118)
- [48] [Compromised AsyncAPI npm Packages: Inside a CI Supply-Chain Attack](https://securitylabs.datadoghq.com/articles/compromised-asyncapi-npm-packages/)
+- [49] [Lightning PyPI compromise: Bun-based credential stealer - Snyk](https://snyk.io/blog/lightning-pypi-compromise-bun-based-credential-stealer/)
+- [50] [Cooldown or compromise: the three-day rule for npm and PyPI - Eye Research](https://research.eye.security/cooldown-or-compromise-the-three-day-rule-for-npm-and-pypi/)
{{#include ../../../banners/hacktricks-training.md}}