Workshop: Teaching Semgrep Assistant to filter out the noise with memories

Mastering Semgrep Memories – Teach Your Assistant to Filter Smarter

False positives slow down security teams, but what if your static analysis tool could learn from your triage decisions? Semgrep Assistant already uses LLMs to filter out findings that purely syntactic engines can’t—but with Semgrep Memories, you can take this even further.

No more writing custom rules for every new library function you consider safe.

No more repeating the same triage steps for each new or refactored repository.

In this hands-on workshop, we’ll walk through how to use human language to teach Semgrep Assistant about trusted data sources, internal sanitizers, and other mitigating context specific to your organization. By leveraging Memories, you ensure that once an issue is triaged, you won’t have to go through the same exercise with similar issues — Assistant learns to recognize the context needed to determine exploitability.

Join us to learn:

✅ A review of how Semgrep Assistant uses contextual analysis and LLMs to filter out the FPs that pure static analysis tools will always flag

✅ The best ways to write Semgrep Memories for accurate filtering

✅ Real-world examples of reducing security noise with Memories

Stop triaging the same issues over and over—let your security tooling learn from you!