Skip to Content
TemplatesEnterprise

enterprise template

Hybrid retrieval with input/output security guards and CI/CD quality gates — for regulated environments and multi-tenant deployments.

What you get

project/
├── src/
│   ├── pipeline.py       # Hybrid pipeline with --input-guard and --output-guard flags
│   └── config.py         # Pipeline configuration
├── eval/
│   ├── golden_set.json   # Evaluation dataset
│   └── config.yaml       # Evaluation thresholds and CI gate settings
├── pyproject.toml        # Python dependencies including rag-forge-core[cohere]
└── README.md

Default configuration

The enterprise template extends the hybrid configuration with input_guard and output_guard enabled. Input guards screen queries for PII and prompt-injection attempts before they reach the retriever. Output guards validate generated answers against the retrieved context before they are returned to the caller.

# pyproject.toml [tool.rag-forge] section
[tool.rag-forge]
template = "enterprise"
chunk_strategy = "recursive"
chunk_size = 512
overlap_ratio = 0.1
vector_db = "qdrant"
embedding_model = "BAAI/bge-m3"
retrieval_strategy = "hybrid"
retrieval_alpha = 0.6
reranker = "none"
enrich = true
input_guard = true
output_guard = true

To invoke with guards active at query time:

rag-forge query "your question" --strategy hybrid --input-guard --output-guard

The CI gate in eval/config.yaml blocks deployments when faithfulness drops below 0.85, giving you a hard quality threshold before any model or data change reaches production.

  1. Index documents with enrichment: rag-forge index --source ./docs --enrich --sparse-index-path .rag-forge/sparse.
  2. Wire the CI gate into your pipeline (GitHub Actions, GitLab CI, etc.) by running rag-forge audit --golden-set eval/golden_set.json as a required check.
  3. Review what the input guard flags on your real query traffic and tune its sensitivity — overly aggressive filtering will degrade user experience.

When to upgrade

The enterprise template is the highest tier in the RAG-Forge stack. There is no template above it. For workflow orchestration needs, the n8n template is a parallel track that can be combined with enterprise-grade pipelines.

Agenticn8n