Model Context Protocol
Add Sgraal as a tool to any MCP-compatible agent host.
{
"mcpServers": {
"sgraal": {
"command": "npx",
"args": ["@sgraal/mcp-server"],
"env": { "SGRAAL_API_KEY": "YOUR_KEY" }
}
}
}
Never let your AI act on unverified memory.
One API call. A deterministic, replayable verdict — the same memory state returns the same verdict, byte-identical, in the absence of feedback events that update internal learning state (strict per-call replay via an opt-in flag).
One wrong memory. One irreversible action. Sgraal flags it before your agent acts on it.
pip install sgraal
content_copy
Design Partner Program
Sgraal's design-partner program is launching across regulated verticals. Partner feedback shapes the roadmap — and identities stay private until launch. Apply to join or contact us.
Anonymized Fortune 500
Healthcare CMO office
Quantitative AI trading team
Confidential
EU-based DPO consortium
Multi-agent platform pilot
Healthcare data network
HIPAA + GDPR cross-jurisdiction
Open-source LLM platform team
Agentic memory governance
US federal contractor
Preflight integration eval
Cards are anonymized to protect partner identity. As partners go public, individual cards are replaced with signed logos and named case studies.
Pick your path
One engine, six customer journeys. Find the page that speaks your language.
Drop-in safety for any AI agent memory. 5-line integration, a Python SDK + MCP server + 6 published framework adapters (3 more in beta), Apache 2.0. SafeMemory wrapper for Mem0 — a code-level integrity gate that catches tampered, forged, or contradicting memory before use.
Read developer docs →Deep dive: Insights · Counterfactual
Run Sgraal at any scale. Hosted, self-hosted, or zero-dep edge mode. Plugin system, autonomous healing, 30-day degradation forecast.
See platform features →Deep dive: OpenTelemetry integration
25 mathematical disciplines. 85 modules. R1–R3 adversarial corpus (available on request). Automated policy-consistency checks. Free academic tier.
Read the whitepaper →Deep dive: Proofs & Certificates
Sell to enterprise faster. Bundle safety + compliance + GDPR proof. W3C VCs, MVMem certificate, founder-accessible pricing.
See the bundle →Open core
No ambiguity. Apache 2.0 SDK and edge mode are open source. The scoring engine is hosted on sgraal.com — your tenant calibration stays private.
pip install sgraal@sgraal/mcpSelf-host the SDK + Proxy + Edge mode for free. The hosted engine powers per-tenant calibration, fleet immunity Beta, and signed compliance certificates.
Every preflight returns USE/WARN/ASK/BLOCK + a one-paragraph explanation. Your agent gets a verdict, your audit log gets evidence.
Byte-identical replay of any historical verdict via cryptographic Memory Passport — in the absence of feedback events that update internal learning state; strict per-call replay via an opt-in flag (see /docs/determinism.md). Forensic-grade audit trail for litigation or compliance.
When the verdict is BLOCK, your agent stops before the irreversible action. Compare to traditional SIEM: detects after the breach.
Sgraal validates every memory access before any irreversible action — even at 3am. No on-call paging for memory-state false positives.
Seamlessly integrates with your existing agent architecture.
{
"headers": {
"Authorization": "Bearer sg_live_..."
},
"body": {
"agent_id": "agent-payments",
"memory_state": [
{"id": "mem_001", "content": "User balance: $50,000",
"type": "semantic", "timestamp_age_days": 3}
],
"action_type": "irreversible",
"domain": "fintech"
}
}
Plain strings in, plain English out. One call.
import sgraal sg = sgraal.Client("sg_live_...") result = sg.check([ "Deploy target is production", "API key is sk-proj-abc123" ]) print(result["safe"]) # False print(result["reason"]) # "Memory contains a likely secret (API key)."
🚀 Beta pricing — all plans are live today. Save 20% with annual billing. Cancel anytime.
Joint benchmark with Grok (xAI) across 5 shared corpora, plus 3 Sgraal-only rounds. Two separately-built stacks scored the same Sgraal-authored corpora side-by-side — convergent, but shared-LLM-prior corroboration, not independent validation.
Recall measured against Sgraal's own ground truth on a synthetic structural corpus (every structurally-detectable attack flagged; ASK_USER counts as caught). Grok independently scored the same corpora — corroboration, not external validation. Residual misses are the metadata-clean content-forgery class (out of scope for a structural gate); over-escalation on benign-control cases is non-zero. Production calibration pending.
60 cases · affiliate bias · brand manipulation
59 cases · commercial_intent 0.30–0.55
60 cases · confident fabrication · multi-hop echo · cross-agent amplification
90 cases · memory injection · cross-agent drift · RAG poisoning · API drift
Timestamp forgery detection. Old decisions disguised as fresh, bypassing Weibull decay.
Gradual role and authority escalation across agent hops.
Self-reinforcing false consensus — no single agent flags the error.
3 independent stacks syncing on fabricated consensus. Joint corpus with Grok — Sgraal-side ready, awaiting partner corpus.
"Confidence ≠ truth. Deterministic structural checks catch what probabilistic systems miss."
R1–R8 structural corpus — every structurally-detectable attack flagged · R14 adversarial hardening in progress
These figures reflect synthetic R12/R14 corpus performance; production calibration is pending paying-customer onboarding.
Joint with Grok across Rounds 1–5 (Sgraal: 100% recall on the structural corpus — every structurally-detectable attack flagged, ASK_USER counts as caught; Grok: F1 0.98–1.000 on the same corpora), plus 3 Sgraal-only rounds (6–8). 614 cases total. Two AI systems stress-tested each other's safety layers as peers. Residual misses are the metadata-clean content-forgery class (out of scope for a structural gate); over-escalation on benign-control cases is non-zero; production calibration pending.
Read the joint blog post →