mirrors/n8n-docs
2
0
Fork 0
mirror of https://github.com/n8n-io/n8n-docs.git synced 2025-11-20 17:48:34 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

Document new Guardrails node in release notes - 1.119.0 (#3853)

Browse Source
This commit is contained in:
Ophir Prusak 2025-11-06 08:57:02 +02:00 committed by GitHub
parent e84efff49d
commit 3e41db00bc
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 13 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

13
docs/release-notes.md
View File

@ -45,6 +45,19 @@ This is the `next` version. n8n recommends using the `latest` version. The `next
This release includes multiple bug fixes for AI Agent, task runners, editor, and integrations, as well as new features like improved workflow settings, AWS Assume Role credentials, and enhanced security and audit capabilities.
<div class="n8n-new-features" markdown>
### Guardrails Node
The Guardrails node provides a set of rules and policies that control an AI agent's behavior by filtering its inputs and outputs. This helps safeguard from malicious input and from generating unsafe or undesirable responses. There are two operations:
- Check Text for Violations: Validate text against a set of policies (e.g. NSFW, prompt injection).
- Sanitize Text: Detects and replaces specific data such as PII, URLs, or secrets with placeholders.
The default presets and prompts are adapted from the open-source [guardrails package](https://github.com/openai/openai-guardrails-js) made available by OpenAI.
<br>
For more info, see [Guardrails documentation](/integrations/builtin/core-nodes/n8n-nodes-langchain.guardrails.md)
</div>
### Contributors
[cesars-gh](https://github.com/cesars-gh)