Self harm guardrail

[jareallen213]

Description

Add form for new self-harm Guardrail (SEMOSS/Semoss#2796)

Adds form configurations for two new embedded guardrail types in the guardrail import flow:

On Topic — rejects prompts whose similarity score against a configured vector database falls below a threshold, with fields for selecting the vector engine, threshold, and nearest-neighbour limit.
Aggressive / Self-Harm — routes prompt evaluation through a configured LLM to detect aggressive or self-harm content, with fields for selecting the model engine and score threshold.

Changes Made

• adds self-harm/aggression guardrail type to guardrail form

How to Test

1. Create new Guardrail
2. Check configured fields + smss file match

Notes

[snyk-io]

✅ Snyk checks have passed. No issues have been found so far.

Status	Scan Engine	Critical	High	Medium	Low	Total (0)
✅	Open Source Security	0	0	0	0	0 issues

💻 Catch issues earlier using the plugins for VS Code, JetBrains IDEs, Visual Studio, and Eclipse.