--- title: "garak trust report" type: "trust-report" slug: "nvidia-garak" canonical_url: "https://www.graphcanon.com/tools/nvidia-garak/trust" computed_at: "2026-07-11T23:42:37.803Z" maintenance_label: "Very active" security_status: "findings" --- # garak trust report _Sourced [trust signals](/glossary/trust-and-signals/trust-signal) from public GitHub metadata and optional dependency [security scans](/glossary/trust-and-signals/security-scan). Not a security guarantee._ ## Maintenance - Label: Very active (96% recency signal) - Days since last push: 1 - Last release: 2026-06-05T17:24:50Z - Releases (90d): 2 - Methodology: github_public_v1 ## Provenance - GitHub repo id: 639097338 - Not a fork - Owner type: Organization - Computed at: 2026-07-11T23:42:37.803Z - Methodology: github_public_v1 ## Security scan - Status: Findings present (0 critical, 0 high, 0 medium, 54 low) · last scan 2026-07-11T23:42:38.356Z ### Findings - **low**: GHSA-29pf-2h5f-8g72 (transformers@5.0 · requirements.txt) - **low**: GHSA-469j-vmhf-r6v7 (nltk@3.9.1 · requirements.txt) - **low**: GHSA-68j8-pq59-fqgm (nltk@3.9.1 · requirements.txt) - **low**: GHSA-7p94-766c-hgjp (nltk@3.9.1 · requirements.txt) - **low**: GHSA-gfwx-w7gr-fvh7 (nltk@3.9.1 · requirements.txt) - **low**: GHSA-h8wq-7xc4-p3qx (nltk@3.9.1 · requirements.txt) - **low**: GHSA-jm6w-m3j8-898g (nltk@3.9.1 · requirements.txt) - **low**: GHSA-p4gq-832x-fm9v (nltk@3.9.1 · requirements.txt) - **low**: GHSA-rf74-v2fm-23pw (nltk@3.9.1 · requirements.txt) - **low**: PYSEC-2026-2078 (nltk@3.9.1 · requirements.txt) - **low**: PYSEC-2026-2085 (nltk@3.9.1 · requirements.txt) - **low**: PYSEC-2026-597 (nltk@3.9.1 · requirements.txt) - **low**: PYSEC-2026-96 (nltk@3.9.1 · requirements.txt) - **low**: PYSEC-2026-97 (nltk@3.9.1 · requirements.txt) - **low**: PYSEC-2026-98 (nltk@3.9.1 · requirements.txt) - **low**: PYSEC-2026-99 (nltk@3.9.1 · requirements.txt) - **low**: GHSA-gr75-jv2w-4656 (langchain@1.2.1 · requirements.txt) - **low**: GHSA-3749-ghw9-m3mg (torch@2.6.0 · requirements.txt) - **low**: GHSA-887c-mr87-cxwp (torch@2.6.0 · requirements.txt) - **low**: GHSA-c678-jfcj-6jmf (torch@2.6.0 · requirements.txt) - **low**: GHSA-f4hp-rmr7-r7v8 (torch@2.6.0 · requirements.txt) - **low**: GHSA-qfhq-4f3w-5fph (torch@2.6.0 · requirements.txt) - **low**: GHSA-rrmf-rvhw-rf47 (torch@2.6.0 · requirements.txt) - **low**: GHSA-vgrw-7cvw-pwgx (torch@2.6.0 · requirements.txt) - **low**: GHSA-x3gm-94wq-g975 (torch@2.6.0 · requirements.txt) - **low**: PYSEC-2025-191 (torch@2.6.0 · requirements.txt) - **low**: PYSEC-2025-198 (torch@2.6.0 · requirements.txt) - **low**: PYSEC-2025-199 (torch@2.6.0 · requirements.txt) - **low**: PYSEC-2025-200 (torch@2.6.0 · requirements.txt) - **low**: PYSEC-2025-201 (torch@2.6.0 · requirements.txt) - **low**: PYSEC-2025-202 (torch@2.6.0 · requirements.txt) - **low**: PYSEC-2025-203 (torch@2.6.0 · requirements.txt) - **low**: PYSEC-2025-204 (torch@2.6.0 · requirements.txt) - **low**: PYSEC-2025-205 (torch@2.6.0 · requirements.txt) - **low**: PYSEC-2025-206 (torch@2.6.0 · requirements.txt) - **low**: PYSEC-2025-207 (torch@2.6.0 · requirements.txt) - **low**: PYSEC-2025-208 (torch@2.6.0 · requirements.txt) - **low**: PYSEC-2025-209 (torch@2.6.0 · requirements.txt) - **low**: PYSEC-2026-139 (torch@2.6.0 · requirements.txt) - **low**: PYSEC-2026-1970 (torch@2.6.0 · requirements.txt) - **low**: GHSA-38vq-g6vr-w8wf (sentencepiece@0.1.99 · requirements.txt) - **low**: PYSEC-2026-1909 (sentencepiece@0.1.99 · requirements.txt) - **low**: GHSA-5wmx-573v-2qwq (markdown@3.4.3 · requirements.txt) - **low**: PYSEC-2026-89 (markdown@3.4.3 · requirements.txt) - **low**: GHSA-cfh3-3jmp-rvhc (pillow@10.4.0 · requirements.txt) - **low**: GHSA-pwv6-vv43-88gr (pillow@10.4.0 · requirements.txt) - **low**: GHSA-r73j-pqj5-w3x7 (pillow@10.4.0 · requirements.txt) - **low**: GHSA-whj4-6x5x-4v2j (pillow@10.4.0 · requirements.txt) - **low**: GHSA-wjx4-4jcj-g98j (pillow@10.4.0 · requirements.txt) - **low**: PYSEC-2026-165 (pillow@10.4.0 · requirements.txt) - **low**: GHSA-6w46-j5rx-g56g (pytest@8.0 · requirements.txt) - **low**: PYSEC-2026-1845 (pytest@8.0 · requirements.txt) - **low**: GHSA-537c-gmf6-5ccf (cryptography@47.0.0 · requirements.txt) - **low**: GHSA-3936-cmfr-pm3m (black@26.1.0 · requirements.txt) ## Method and caveats These are heuristics from public GitHub data and optional dependency scans, sourced and dated. "No criticals found on 2026-07-11" is not a guarantee of safety. See the full methodology: [/trust-methodology](/trust-methodology.md). ## Common questions ### Is garak maintained? GraphCanon rates garak "Very active" (96% maintenance signal from public GitHub metadata, computed today). Last push was 1 days ago. 2 GitHub releases in the last 90 days. This is a recency heuristic, not a guarantee the project will stay maintained. ### Is garak safe to use? Last scanned today (deps profile). Status: 54 low - 54 low finding(s) in the latest scan. GraphCanon does not claim the project is safe or vulnerability-free; review findings on the trust report. GraphCanon does not certify garak as safe - review maintenance, provenance, and scan findings on this page before adopting. ### Is garak a fork? No. garak is not flagged as a fork in GitHub metadata at the time of the last refresh. ### Does garak have known security vulnerabilities? Last scanned today (deps profile). Status: 54 low - 54 low finding(s) in the latest scan. GraphCanon does not claim the project is safe or vulnerability-free; review findings on the trust report. ### How often is the garak trust report updated? Trust signals refresh on GitHub ingest/refresh cycles and optional dependency/MCP scans. This report was computed today (methodology github_public_v1). ### What does GraphCanon never claim about garak? We never publish a composite safety grade, pen-test endorsement, or "verified secure" label for garak. Signals are sourced heuristics with explicit limits - see [trust methodology](/trust-methodology). ### How does GraphCanon assess trust for garak? Signals are sourced from public GitHub metadata and optional dependency/MCP manifest scans, each tagged with methodology version and computed date. GraphCanon does not publish a composite safety grade. Read [trust methodology](/trust-methodology) for full scope and limits. --- **Machine-readable endpoints** - JSON: [`/api/graphcanon/tools/nvidia-garak/trust`](/api/graphcanon/tools/nvidia-garak/trust) - LLM index: [/llms.txt](/llms.txt) - Full corpus: [/llms-full.txt](/llms-full.txt) _GraphCanon - The knowledge graph for AI development. https://www.graphcanon.com/_