Loading the catalogue…
Chat UIs · Mintplex Labs Inc. · US
AnythingLLM
US CLOUD Act exposureOpen sourceMIT
STAV REVIEW · LAST REVIEWED 06/21/2026
“Genuinely Sovereign Inference architecture with broad deployment flexibility, but 20 security advisories in under two years — including a CVSS 9.8 RCE — and a privacy policy that has been offline for seven weeks demand serious compensating controls.”
Good for
- Teams building internal RAG pipelines who want Inference routed directly to their own LLM backend with no vendor proxy in the data path
- Air-gapped or EU-regulated deployments where prompt data must stay on operator-controlled infrastructure, paired with a local Inference backend and DISABLE_TELEMETRY=true enforced
- Organisations already running Docker or bare-metal servers who want a full-featured document-workspace UI without a SaaS dependency
- Teams who want to channel all Inference through the Stav Smart Router for audit-logged oversight of an otherwise self-managed stack
- Open-source-first buyers who need MIT licence terms with no copyleft obligations and a documented admin API for custom integrations
Watch out for
- Security advisory velocity is structural, not incidental: 20 published GHSAs total, 7 in the past 90 days, peak CVSS 9.8 (XSS-to-RCE via LLM response injection) — all individually resolved, but the cadence requires active release pinning and a subscribed GHSA advisory feed
- The vendor's privacy policy has returned HTTP 404 across six or more consecutive curator runs spanning over seven weeks; telemetry behaviour is attested only in a project README with no formal legal backing — enforce DISABLE_TELEMETRY=true at the environment level before any team activates the app
- No SSO, no SOC 2, no ISO 27001, no DPA, no bug bounty programme, unsigned binaries, no SBOM, and a bus factor of 1 on the primary maintainer — unsuitable for strictly regulated enterprise deployments without documented compensating controls
Editorial assessment, not legal advice. Stav's risk ratings, scores, and verdicts are our own analysis of publicly available information and may be incomplete or out of date. Verify independently before making compliance or procurement decisions.
MODEL REQUIREMENTS
What AnythingLLM needs from a model.
Required
- Streaming
Recommended
- Document understanding
- Long context (≥128k)
- Structured output
- Tool calling
Known issues8 open
Public security, privacy and governance advisories that are still open. Each row carries a source link — these are factual claims, not editorial judgements.
2026-06-30
lowUpstream change on anythingllm: github_release 'v1.14.2' → 'v1.15.0', github_last_push '2026-06-23T00:03:53Z' → '2026-06-30T04:24:07Z'
other · open
Source ↗2026-06-23
lowUpstream change on anythingllm: github_release 'v1.14.0' → 'v1.14.2', github_last_push '2026-06-16T05:00:01Z' → '2026-06-23T00:03:53Z'
other · open
Source ↗2026-06-16
lowUpstream change on anythingllm: github_release 'v1.13.0' → 'v1.14.0', github_last_push '2026-06-09T04:01:04Z' → '2026-06-16T05:00:01Z', cve_count 19 → 20
other · open
Source ↗2026-06-09
lowUpstream change on anythingllm: github_last_push '2026-06-02T04:28:00Z' → '2026-06-09T04:01:04Z'
other · open
Source ↗2026-06-02
lowUpstream change on anythingllm: github_release 'v1.12.1' → 'v1.13.0', github_last_push '2026-05-19T01:51:34Z' → '2026-06-02T04:28:00Z', cve_count 16 → 19