Company profile
Escape
Agentic offensive AppSec: continuous discovery, AI pentesting (Cascade) and remediation in engineering workflows, now covering AI chatbots, agents and the MCP servers behind them.
- Status
- PrivateListed on CyberMerge Startup Radar (private vendors), snapshot Nov 13, 2026
- Valuation
- Undisclosed
- Latest funding
- Undisclosed
- HQ
- Paris / France (Startup Radar)
- Last updated
Categories
Funding & valuation
Funding: Undisclosed — no round recorded in the CyberMerge Funding Tracker.
Latest signals
- Startup Radar · Application Security · snapshot Nov 13, 2026 21 Oct 2026: launched AI Red Teaming inside its AI Pentesting product, live for customers. Cascade attacks an MCP server through the chatbot that fronts it, or pentests the MCP server directly by enumerating its tools over MCP's JSON-RPC protocol; it holds multi-turn jailbreak conversations and only reports a prompt injection when it proves a real result (an unauthorized tool call, another tenant's data, a privileged action), all in the same pentest as the web app (Escape blog). Last round: $18M Series A led by Balderton (Mar 2026); total ~$23M; valuation Undisclosed. Source
Competitors
Same-category peers from CyberMerge Research maps, Startup Radar and What's The Use tags.
- ArmadinPrivateAI-native offensive security: autonomous agent swarms chain low-severity weaknesses into validated kill…
- OX SecurityPrivateAI-native AppSec from prompt to runtime: ASPM/PBOM, VibeSec for AI-generated code, plus OX Cloud CNAPP with…
- SnykPrivateDeveloper-first AppSec platform: code (SAST), open-source dependency (SCA), container and IaC scanning, now…
- HadrianPrivateAgentic offensive security: Atlas continuously maps the external attack surface and uses AI agents to…
- SemgrepPrivateOSS-rooted AppSec platform (SAST, SCA, secrets), now rebuilt so coding agents can find, triage and fix…
- Aikido SecurityPrivateUnified developer AppSec suite (SAST/SCA/secrets/cloud/runtime) with autofix; European unicorn.
- ApiiroPrivateDeep-code ASPM with Risk Graph, design-phase AI threat modeling, AutoFix.
- ArmorCodePrivateUnified exposure / ASPM orchestration across AppSec, cloud, and AI findings.
- ChainguardPrivateSafe-source open-source: zero-CVE container images, libraries, and supply-chain factory.
- CodeWallPrivateAutonomous AI offensive security: multiple specialized agents start with no credentials or inside knowledge,…
- Fleuret AIPrivateAgentic continuous pentesting: two AI agents (Emile and Champollion) map exposed apps, APIs and…
- GitGuardianPrivateSecrets detection + non-human identity / AI-agent credential governance across SDLC.
Figures appear only as recorded in CyberMerge data, with source and date. “Undisclosed” means no published figure in our data. Editorial, not investment advice. Last updated . Spot an error? Tell the desk.
