Pillar Security
by Pillar Security
AI security platform for discovering, red-teaming and guarding AI agents, MCP servers and coding agents, being acquired by Vanta
Pillar Security is an AI security platform that inventories an enterprise's AI agents, models, MCP servers and coding assistants, red-teams them with autonomous attacks and enforces runtime guardrails. It is for CISOs and AppSec teams who need evidence that agentic AI is tested and governed. Vanta signed an agreement to acquire it on November 9, 2026.
Pillar Security was founded in 2023 by Dor Sarig and Ziv Karliner, with offices in Tel Aviv and Miami, and raised a $9M seed round in April 2025 led by Shield Capital with Golden Ventures and Ground Up Ventures. Its platform covers the AI lifecycle in five modules. AI Discovery and Posture catalogs agents, models, MCP servers, skills and coding agents without an agent install, including shadow AI, and scans their configs, permissions and identities. Red teaming runs through RedGraph and the Red Graph Suite launched August 6, 2026, which maps an agent's system prompt, tools and data sources, runs multi-turn attacks through the live application's interface, records video and logs as evidence, and tracks attack success rate release over release. Runtime Guardrails mask or block PII, PHI and secrets and monitor tool calls. Pillar for AI Coding Agents, launched February 5, 2026, scans .cursor and .claude configuration directories on workstations for risky MCP connections, auto-run flags and hardcoded credentials across Claude Code, Cursor, Codex, GitHub Copilot and Antigravity. Findings map to OWASP's LLM and Agentic Top 10, MITRE ATLAS, NIST AI RMF, the EU AI Act and Pillar's own SAIL framework. Named customers include Similarweb, Eleos Health, AvidXchange and Tavily. Gartner named it a 2026 Cool Vendor in AI software security. On November 9, 2026 Vanta signed a definitive agreement to buy Pillar, reported by Calcalist at about $150M in cash and shares, and will roll its capabilities into Vanta in stages through summer 2027.
A CISO or AppSec lead who must prove to auditors that agentic AI apps and developer coding agents are inventoried, attack-tested and policed at runtime
One inventory of AI assets plus reproducible red-team evidence mapped to OWASP, MITRE ATLAS, NIST AI RMF and the EU AI Act
At a Glance
- Category
- Governance & Security
- Pricing
- Contact for pricing
- Target Market
- CISOs, AppSec Teams, AI Platform Teams, GRC Leaders, CTOs
- Deployment
- Cloud-first, Self-hosted, Hybrid
- Founded
- 2023
- Headquarters
- Tel Aviv, Israel
- Team Size
- 11-50
Key Features
- ✓AI Discovery and Posture (AI-SPM)
Agentless inventory of agents, models, MCP servers, skills and coding agents, including shadow AI, with config and permission scanning.
- ✓Red Graph Suite
Autonomous multi-turn attacks through the live app interface, with video and logs as evidence and per-release posture tracking.
- ✓Runtime Guardrails
Adaptive guardrails that mask, block or log PII, PHI, credentials and secrets and monitor agent tool calls in production.
- ✓Pillar for AI Coding Agents
Workstation scanning of Claude Code, Cursor, Codex and Copilot configs for untrusted MCP servers, auto-run flags and hardcoded credentials.
- ✓Framework-mapped governance
Turns policies into controls and produces audit reports mapped to OWASP, MITRE ATLAS, NIST AI RMF and the EU AI Act.
- ✓Flexible deployment
Cloud, self-hosted and hybrid options, including deployment in the customer's own VPC with its own data and models.
Use Cases
- •Pre-release agent testing
AppSec runs autonomous attacks against a new support agent and blocks release when attack success rates regress.
- •Coding-agent hygiene
Security teams scan developer machines for MCP servers bypassing the AI gateway and remove hardcoded credentials from configs.
- •Regulated data protection
Healthcare and fintech teams mask PHI and PII in LLM traffic at runtime and keep logs for auditors.
- •AI audit readiness
GRC teams export findings mapped to the EU AI Act and NIST AI RMF to show auditors that AI risk is controlled.
Ideal For
Best For
- ✓Continuous red-teaming of customer-facing AI agents before and after each release
- ✓Finding shadow AI and risky MCP server connections on developer workstations
- ✓Runtime masking of PII, PHI and secrets in LLM applications
- ✓Producing audit evidence mapped to OWASP Agentic Top 10, NIST AI RMF and the EU AI Act
- ✓Existing Vanta customers building an AI governance program
Not Ideal For
- ✗Buyers who need a long-term standalone roadmap: the Vanta deal is signed but not closed, and neither company has said how existing Pillar contracts, pricing or support will be handled
- ✗Teams that want published self-serve pricing or a free tier, since none is public
- ✗Organisations needing a large vendor with a long track record: Pillar has about 20 employees and was founded in 2023
Deployment
Market Analysis
Pros
- ✓Broad coverage across discovery, red teaming, runtime guardrails and governance in one platform
- ✓Research team has disclosed patched flaws in Cursor, Codex and Gemini CLI (per Calcalist)
- ✓SOC 2 Type II and ISO 27001, with in-VPC deployment
- ✓Recognised as a 2026 Gartner Cool Vendor in AI software security
Cons
- ✗Acquisition uncertainty: Channel Insider notes pricing, packaging and licensing under Vanta are unknown
- ✗Runtime enforcement inside Vanta is not due until summer 2027, so early Vanta offerings are detection and monitoring
- ✗Small company (about 20 staff per Calcalist) with no public pricing and no independent review-site ratings found
Pricing
Enterprise
Contact for pricing
- ✓AI discovery and posture
- ✓Red teaming (RedGraph)
- ✓Runtime guardrails
- ✓Governance and compliance reporting
Pillar publishes no list pricing; sales are quote-based (pillar.security/pricing returned 404 on November 14, 2026). After the Vanta deal, Vanta says AI Governance becomes a paid Vanta offering in January 2027, adversarial testing follows in spring 2027 and runtime enforcement by summer 2027, but packaging and price for those have not been announced.
Security & Compliance
Sources
This page was written from 7 sources, 4 on domains other than pillar.security.
- 1.pillar.security — pillar.securityvendor
- 2.pillar.security — pillar launches red graph suite version controlled contextuavendor
- 3.pillar.security — introducing pillar for ai coding agentsvendor
- 4.calcalistech.com — hkjzjkjrfl
- 5.channelinsider.com — news vanta pillar security deal ai protection
- 6.refreshmiami.com — pillar security raises 9m to create ai security guardrails f
- 7.hn.algolia.com — search
Stay Ahead of the Curve
Weekly enterprise AI insights for technology leaders. No spam, no vendor pitches—unsubscribe anytime.
SubscribeRelated Products
Bay
AI agent security platform that discovers every agent and MCP server and blocks risky agent actions at runtime
Larridin
AI value measurement platform that tracks enterprise AI spend, adoption and output across human and agent work
Cyara Agentic CX Trust Platform
Independent testing, monitoring and observability for AI agents in voice and digital customer service
Sysdig AI Defense
Runtime security and governance that inventories AI agents and blocks risky actions before they complete