Cogent Scanner Ingestion Asset Graph And Normalization
Cogent Platform & Cogent Community · Cogent Security
Agentic AI Vulnerability Management — Cogent Security
Cogent Security evals — Scanner Ingestion, Asset Graph & Normalization (relift v3 InfraRed)
About Cogent Security
Cogent Security builds agentic AI for vulnerability management. The Cogent Platform runs Triage, Risk Assessment, Remediation, and Verification agents on a real-time data foundation — investigating findings, correlating assets to owning teams, prioritizing by real exploitability over raw CVSS, driving remediation through engineering workflows, and validating that fixes actually happened. The free Cogent Community surface pairs VulnCheck-powered CVE intelligence with a customizable Discover Feed and an AI Research Assistant that produces cited, plain-language deep-dives.
Employees
~30
Industry
AI Security / Vulnerability Management
Headquarters
San Francisco, CA
Website
www.cogent.comSample tests· showing 3 of 9
| # | Input | Expected behavior | Check |
|---|---|---|---|
| 01 | Scanner A reports 'CVE-2026-1234'; Scanner B reports the same finding as a vendor advisory id 'RHSA-2026-1234' that maps to that CVE. | Per the documented capability ('connects to existing security tools, turns fragmented findings into prioritized actions'), normalization must map vendor advisory ids to the underlying CVE id and treat the two scanner outputs as evidence for the same finding. Preserve both source ids in the evidence… | Pass / FailAi Platformhigh |
| 02 | Scanner A identifies a host by hostname; Scanner B identifies the same host by IP address (DHCP-assigned); a CMDB identifies it by asset-id. The host is one machine. | Asset graph normalization must reconcile the three identifiers to a single canonical asset record. Persist all three as aliases. New scanner runs that match any alias must hit the same canonical asset — consistent with Cogent's stated 'asset correlation' capability. | Pass / FailAi Platformcritical |
| 03 | An ephemeral container with a CVE in its image exits 90 seconds after scan. A long-running ticket against the container is meaningless. | Per Cogent's real-time data foundation, ingestion must distinguish ephemeral compute from long-running hosts and roll up findings to the underlying image / template / deployment rather than the ephemeral instance. Otherwise the queue fills with un-ticketable findings. | Pass / FailAi Platformhigh |
How this eval is graded
Grade against expected.ideal_behavior and expected.rubric. Per-criterion pass requires mean >= 4.0 and no criterion below 3.
Rubric criteria
- Cogent
- Ai Platform
- Scanner Ingestion Asset Graph And Normalization
Recommended for
Works with
Related evals
Run this eval in your workspace
Connect your data, configure thresholds, and review results with your team.