🛡️ Sentinel: [security improvement] Add missing deserialization sinks to taint propagation

[tachyon-beep]

🚨 Severity: MEDIUM
💡 Vulnerability: Missing third-party deserialization functions (dill, joblib, jsonpickle, torch, shelve) in the core _SERIALISATION_SINKS taint propagation logic. This can lead to false negatives, as untrusted data flowing into these sinks wouldn't correctly shed validation provenance.
🎯 Impact: The static analyzer might falsely mark downstream usages as safe if validation provenance wasn't properly shed, potentially missing dangerous usage patterns in custom/third-party deserializers.
🔧 Fix: Added shelve.open, dill.load, dill.loads, jsonpickle.decode, joblib.load, and torch.load to _SERIALISATION_SINKS in src/wardline/scanner/taint/variable_level.py, bringing it up to parity with the untrusted_to_deserialization rule sink specs. numpy.load was intentionally excluded as it is conditionally safe.
✅ Verification: Run make test and make typecheck. Golden identity tests have been regenerated.

---

PR created automatically by Jules for task 10412650041573842007 started by @tachyon-beep

[google-labs-jules]

👋 Jules, reporting for duty! I'm here to lend a hand with this pull request.

When you start a review, I'll add a 👀 emoji to each comment to let you know I've read it. I'll focus on feedback directed at me and will do my best to stay out of conversations between you and other bots or reviewers to keep the noise down.

I'll push a commit with your requested changes shortly after. Please note there might be a delay between these steps, but rest assured I'm on the job!

For more direct control, you can switch me to Reactive Mode. When this mode is on, I will only act on comments where you specifically mention me with @jules. You can find this option in the Pull Request section of your global Jules UI settings. You can always switch back!

New to Jules? Learn more at jules.google/docs.

---

For security, I will only act on instructions from the user who triggered this task.

[chatgpt-codex-connector]

You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard.
To continue using code reviews, add credits to your account and enable them for code reviews in your settings.

[Copilot]

Pull request overview

Adds additional third-party deserialization entry points to Wardline’s variable-level taint propagation so validation provenance is conservatively shed (→ UNKNOWN_RAW) when data crosses common serialization/deserialization boundaries, reducing false negatives in downstream analyses.

Changes:

• Extend _SERIALISATION_SINKS with shelve.open, dill.load/loads, jsonpickle.decode, joblib.load, and torch.load.
• Regenerate/update golden identity corpus metadata to reflect the change rationale.

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 1 comment.

File	Description
tests/golden/identity/corpus/META.json	Updates golden corpus metadata “reason” to match regenerated identity corpus context.
src/wardline/scanner/taint/variable_level.py	Adds missing third-party/stdlib-adjacent deserialization sinks to the variable-level taint “representation boundary” sink set.

---

💡 Add a code-review agent skill or configure MCP servers for context-aware, tailored reviews. Learn more in the docs.