Registry › Compare › Garak vs Guardrails AI

Garak vs Guardrails AI

Guardrails AI leads on trust: 76.4/100 (Grade B) against 65.6/100 (Grade B), a 10.8-point gap. Garak leads on maintenance; Guardrails AI leads on supply-chain integrity and provenance.

B Garak 65.6

the LLM vulnerability scanner

NVIDIA/garak · #476 overall · #7 Security & Guardrails · coverage A (4/5)

Choose Garak if maintenance matters most.

+6.3Maintenance: last push 3d ago, against 4d ago

B Guardrails AI 76.4

Adding guardrails to large language models.

guardrails-ai/guardrails · #218 overall · #5 Security & Guardrails · coverage A (4/5)

Choose Guardrails AI if supply-chain integrity and provenance matter most.

+10.4Safety / Integrity: 100% of recent commits signed, against 72%
+7.2Identity / Provenance: package provenance attested, against none
+1.0Transparency: OSSF Scorecard 6.0 against 4.8

Where they differ

9.6

Safety / IntegrityGuardrails AI +10.4

20.0

10.8

Identity / ProvenanceGuardrails AI +7.2

18.0

12.6

TransparencyGuardrails AI +1.0

13.6

18.0

MaintenanceGarak +6.3

11.7

14.9

AdoptionGuardrails AI +0.2

15.1

-0.3

Runtime calibrationGarak +1.7

-2.0

Full evidence table

An independent, evidence-based trust comparison of Garak and Guardrails AI, two Security & Guardrails projects in the HVTracker registry. Scores come from public, checkable signals — supply-chain provenance, OSSF Scorecard, maintenance, and adoption — not popularity.

Full evidence

Signal	GarakNVIDIA/garak	Guardrails AIguardrails-ai/guardrails
HVTrust score	65.6	76.4
Evidence grade	B	B
Coverage grade	A	A
Overall rank	#476	#218
Rank in Security & Guardrails	#7	#5
GitHub stars	9.5k	7.5k
Last updated	3d ago	4d ago
Build provenance	No	Yes
OSSF Scorecard	4.8 / 10	6.0 / 10
License	Apache-2.0	Apache-2.0
Downloads	10k/wk	24k/wk
Trust dimensions (points earned)
Safety / integrity / 25	9.6	20.0
Identity & provenance / 18	10.8	18.0
Transparency / 17	12.6	13.6
Maintenance / 20	18.0	11.7
Adoption / 20	14.9	15.1
Runtime capability surface (full matrix)
MCP server	—	—
External providers	8 — Amazon Bedrock, Anthropic, Cohere, …	3 — Anthropic, Multi-provider (LiteLLM), OpenAI
Requires API keys	Yes	Yes
Plugin surface	plugins	—
Provenance drift	Match	Unknown

Open in the live compare tool → Garak profile Guardrails AI profile More Security & Guardrails →

How to read this: HVTrust (0–100) weighs supply-chain signals (provenance, OSSF Scorecard, signed commits, open license) alongside real-world adoption, scaled by an evidence-confidence factor. Grade bands: A ≥ 80, B ≥ 65, C ≥ 50, D < 50. Signals refresh daily. Full methodology v4.4 →