================================================================================
SAFEGUARD PRESS KIT
Software Supply Chain Security — AI-Native, Self-Healing
================================================================================

COMPANY BOILERPLATE
-------------------
Safeguard is the AI-native software supply chain security platform.
Safeguard catches Zero Days, autonomously remediates them at 100-layer
dependency depth, and ships 500K+ curated zero-CVE components so customers
deploy clean. The platform serves regulated enterprises, AI-forward platform
teams, critical infrastructure, medical device manufacturers, scale-ups under
SOC 2, and M&A acquirers across the United States, India, EMEA, and APAC.

Founded in 2024. Headquartered in Dublin, California.
Email: press@safeguard.sh · hi@safeguard.sh
Website: https://safeguard.sh

================================================================================

KEY MESSAGING (one-liners for headlines)
----------------------------------------
- "Safeguard & Self-Heal your software supply chain."
- "500K+ zero-CVE components. 50+ Zero Days discovered. 100K+ AI remediations
   applied. 1M+ scans completed."
- "AI-Native. Zero Day discovery. Autonomous remediation."
- "100-layer reachability — 40 layers deeper than legacy SCA."
- "Defense in depth, in four concentric layers, around every customer asset."

================================================================================

PLATFORM STATS (point-in-time totals, as of June 2026)
------------------------------------------------------
- 500K+   curated zero-CVE components across npm, PyPI, Maven, NuGet, Go,
          Rust, RubyGems, PHP, containers, and Helm
- 50+     Zero Days discovered via Griffin AI reachability + taint analysis
- 100K+   autonomous AI remediations applied
- 1M+     cumulative scans completed
- ~3 days mean time to remediate (vs typical industry baselines of weeks to months)
- ~92%    faster remediation vs traditional SCA workflows (typical)
- ~80%    fewer false positives via reachability analysis (typical)

================================================================================

PRODUCTS
--------
- Griffin AI            — autonomous security agent, 100-layer reachability
- Attention Architecture — model serving infrastructure for the Griffin lineup
- ESSCM                 — Enterprise Software Supply Chain Manager
- SBOM Studio           — ingest, slice, distribute CycloneDX + SPDX
- Scanner Suite         — 11 integrated scanners + enrichment feeds
- TPRM                  — Third Party Risk Manager
- Open Source Manager   — registry intelligence for OSS dependencies
- Auto-Fix              — autonomous remediation engine
- MCP Server            — agent surface for Claude Code / Cursor / Cline
- IaC Security          — Terraform / Pulumi / Crossplane policy gates
- DAST                  — dynamic application security testing
- Safeguard Code        — local agent for IDE / CLI
- Safeguard Guard       — runtime protection layer
- Safeguard Cowork      — collaboration surface for security teams
- Marketplace           — Gold Registry of zero-CVE artifacts

GRIFFIN MODEL FAMILY
--------------------
- Griffin Lite (8B)     — on-prem inference, edge deployment
- Griffin S (14B)       — small-team triage workloads
- Griffin M (32B)       — standard enterprise tier
- Griffin L (70B)       — high-throughput, reachability + remediation
- Griffin Zero (671B-MoE) — sovereign tier, deep call-graph context

OTHER MODELS
- Eagle (13B)           — ranking + clustering, taint-path benchmarks
- Lino (1B)             — distilled inline model, on-device latency

================================================================================

COMPLIANCE POSTURE
------------------
- SOC 2 Type II (audit in progress)
- ISO/IEC 27001:2022 (aligned)
- Designed for FedRAMP HIGH · IL5 GovCloud variant
- CMMC L2/L3 mappings for DoD supply chain
- DORA + NIS2 control mappings (EU)
- DPDP Act mapping (India) · GDPR mapping (EU)
- EO 14028 / NIST SSDF self-attestation aligned

================================================================================

REGIONAL FOOTPRINT
------------------
Global  — app.safeguard.sh (GA default endpoint)

United States (regional tenants):
- US East (Virginia · Ashburn) · US East (Ohio · Columbus)
- US West (California · San Jose) · US West (Oregon · Portland)
- US Central (Texas · Dallas) · US South (Georgia · Atlanta)
- US Northeast (New York) · US GovCloud (FedRAMP High · IL5)

India (full domestic coverage):
- GIFT City (Gujarat · IFSCA · sovereign-tier financial)
- Mumbai · Hyderabad · Bengaluru · Chennai · Visakhapatnam
- Delhi NCR · Pune · Kolkata
- India Sovereign (MeitY · CERT-In · STQC)

Europe: Frankfurt · Paris · Amsterdam · Stockholm · Madrid · Milan ·
        Warsaw · London · Zurich · Dublin

Middle East: Dubai · Riyadh · Manama · Tel Aviv · Doha
Asia-Pacific: Singapore · Tokyo · Osaka · Seoul · Hong Kong · Sydney ·
              Melbourne · Auckland · Jakarta · Kuala Lumpur · Manila ·
              Bangkok · Ho Chi Minh

================================================================================

MEDIA CONTACTS
--------------
Press inquiries:     press@safeguard.sh
General contact:     hi@safeguard.sh
Sales:               sales@safeguard.sh
Documentation:       docs@safeguard.sh
Trust Center:        https://safeguard.sh/security

Brand assets (logos, screenshots, video):
- Request via press@safeguard.sh — full kit shared as an authenticated link.

LinkedIn: https://www.linkedin.com/company/safeguard-sh
X / Twitter: https://x.com/safeguard_sh
GitHub: https://github.com/Safeguard-sh
NPM: https://www.npmjs.com/org/safeguard-sh

================================================================================

USAGE NOTES FOR JOURNALISTS
---------------------------
- Capitalise the brand as "Safeguard". Not "safeguard" or "SafeGuard".
- The product's AI agent is "Griffin AI" (always two words).
- Quote attribution: please reach out before quoting executives so we can
  confirm the latest title and verify the quote in context.
- The "500K+", "50+", "100K+", "1M+" figures are point-in-time totals updated
  monthly. The current snapshot was taken June 2026. Drill-down breakdowns
  are published at https://safeguard.sh/stats/<metric>.

================================================================================
Last updated: June 2026
