Cisco open sources Foundry Security Spec to make AI cybersecurity findings verifiable and auditable at machine speed

Published on 15 May 2026

Its “done” signals aim to stop AI from freelancing

Cisco has open-sourced the Foundry Security Spec, an AI security evaluation blueprint designed to replace noisy, unverifiable alerts with structured, auditable findings. Built for machine-speed threats, it counters frontier-model hallucinations using orchestration, bounded outputs, and clear completion signals. The spec is model-agnostic and stack-neutral, and it’s released as two artifacts: Spec.md with ~130 requirements and Constitution.md with 11 inviolable principles tied to real failures Cisco encountered.

Foundry shifts security evaluation from alerts to structured, auditable findings
It uses agent roles and guardrails to keep AI outputs bounded and prioritized
Published as Spec.md and Constitution.md, totaling about 130 functional requirements plus principles
The approach is model-agnostic and stack-neutral, working across different LLMs and infrastructure
Defenders can plug into GitHub spec-kit workflows for spec-driven evaluation
Cisco says it is not a turnkey scanner and requires human governance and local adaptation

#open source #cybersecurity #cisco #llm governance #ai security

Read the full story at IndianWeb2

This summarization was done by Beige for a story published on IndianWeb2

Cisco open sources Foundry Security Spec to make AI cybersecurity findings verifiable and auditable at machine speed

The full experience is on mobile.