Public story · 2026-02-23 · source-backed
Cisco Open-Sources A2A Protocol Scanner
Story
The first security tool specifically for Google's Agent-to-Agent protocol. Five detection engines: pattern matching, protocol validation, behavioral analysis, runtime endpoint testing, and LLM-powered semantic analysis. Covers 17 threat types including agent impersonation and prompt injection via Agent Cards. MIT licensed.
Ironic twist: CVE-2026-26057 — Cisco's own skill-scanner API server had a vulnerability allowing DoS and arbitrary file upload (patched in 1.0.2). The "who watches the watchmen" problem is real.
Source: Cisco Blog | GitHub
Related stories
Each link below shares sources, entities, or timing with this story.
Shared entities / Same source domain / Shared topic / What happened next
A Markdown File Got 44K Stars This Week. Skills Are the New Package Manager.
Both cover GitHub, Google, LLM; reported by the same outlet (github.com); overlapping topics (agent, behavioral, file, google).
Shared entities / Same source domain / Shared topic / What happened next / Tension
Cisco Ships Four Open-Source Agent Security Scanners
Both cover Cisco, Cisco Blog; reported by the same outlet (blogs.cisco.com); overlapping topics (agent, analysi, cisco, file).
Shared entities / Same source domain / Shared topic / What happened next / Downstream implication
Agent-threat-rules ships "Sigma for AI agents": 425 open detection rules, 97.1% recall, already in Microsoft AGT, Cisco AI Defense, MISP and OWASP.
Both cover Agent, GitHub; reported by the same outlet (github.com); overlapping topics (agent, cisco, detection).
Shared entities / Same source domain / Shared topic / What happened next
Cisco MCP Scanner — Enterprise MCP Security (1.2K+ stars)
Both cover GitHub, LLM; reported by the same outlet (github.com); overlapping topics (analysi, behavioral, cisco, engin).
Shared entities / Same source domain / Shared topic / What happened next / Tension
promptfoo — LLM Red Teaming (+718 stars/day, 12.5K total)
Both cover GitHub, LLM; reported by the same outlet (github.com); overlapping topics (agent, cover).
Shared entities / Same source domain / Shared topic / What happened next
DeepAudit v3.0 — Multi-Agent Vulnerability Detection (4.8K stars)
Both cover GitHub, LLM; reported by the same outlet (github.com); overlapping topics (agent, analysi, detection).
Shared entities / Same source domain / Shared topic / Earlier coverage / Tension
garagon/aguara — Static Security Scanner for Agent Skills (16 stars, 3 days old)
Both cover GitHub, LLM; reported by the same outlet (github.com); overlapping topics (agent, analysi).
Shared entities / Same source domain / What happened next
Agent memory is the most contested OSS category of mid-2026.
Both cover Agent, GitHub, LLM; reported by the same outlet (github.com); picks up the Agent thread on 2026-06-07.
Source trail
Entities
Provenance
- Canonical issue
- Ramsay Research Agent — 2026-02-23
- AI generated
- no
- Story unit
- 2026-02-23-cisco-open-sources-a2a-protocol-scanner
- Labels
- source-backed, canonical briefing excerpt