Sources

OWASP Top 10 for Agentic Applications (Dec 9 2025); OWASP threats/mitigations (Feb 2025) and securing guide 1.0 (Jul 2025); MITRE ATLAS agent techniques (Oct 2025, with Zenity); Microsoft failure-mode taxonomy v2.0 (Jun 4 2026); CISA + five partners, Careful Adoption of Agentic AI Services (May 1 2026, verified against https://content.govdelivery.com/accounts/USDHSCISA/bulletins/41544ff and https://www.cyber.gov.au/business-government/secure-design/artificial-intelligence/careful-adoption-of-agentic-ai-services); NIST NCCoE agent identity concept paper (Feb 5 2026); NIST SP 800-207; Entra ID Protection docs [vendor]; Visa Trusted Agent Protocol (Oct 2025) [vendor]; Mastercard Agent Pay (Apr 29 2025) [vendor]; Google AP2 (Sep 16 2025) [vendor]; AP2 red-team (arXiv 2601.22569, Jan 2026); PSD3/SCA analyses (Norton Rose, BRC, 2026); IEC 61511 AI exclusion (via Kenexis); ISO/IEC TR 5469:2024; FINRA 24-09 (Jun 27 2024) and FINRA 25-07 (Apr 14 2025, request for comment, not a final interpretation), plus the 2026 oversight report; ESMA statement (May 2024); Deloitte Australia refund (Fortune, Oct 2025); Bhattarai & Vu deterministic boundaries (arXiv 2602.09947, Feb 2026); guardrail bypass (arXiv 2504.11168, ACL LLMSEC 2025; arXiv 2502.15427; arXiv 2605.06669; arXiv 2410.02916); FIDES (arXiv 2505.23643; Agent Framework release) [vendor]; Bedrock Automated Reasoning GA (Aug 6 2025) [vendor]; policy-engine guides (Oso 2025, Natoma 2026 [vendor]); EchoLeak (Dark Reading, Jun 2025); Lattice reversal (Fortune, Jul 2024); Agent 365 licensing (SAMexpert 2026; Microsoft GA post May 1 2026 [vendor]; ~40M agents: FY26 Q4 earnings self-report); CSA/Oasis NHI survey (Jan 2026 [sponsored]); Entro NHI stats [vendor]; acquisitions: CSO Online, SecurityWeek, Infosecurity, Cremit RSAC 2026; Gartner guardian agents (Jun 11 2025) + Market Guide (Feb 2026, via reprints); Microsoft least-privilege for agents (Jul 16 2026) [vendor]; OAuth RAR RFC 9396 guides. Re-verify by Q4 2026: consolidation state; Agent 365 registration figures; SIEM baselining gap; guardrail benchmark updates.

Added 2026-10-14 (accessed 12 to 14 October 2026): Microsoft Agent Framework security documentation, FIDES marked experimental (7 Oct 2026) https://learn.microsoft.com/en-us/agent-framework/agents/security [vendor]; Nasr, Carlini et al., adaptive attacks against prompt-injection defences, arXiv 2510.09023 (Oct 2025); NIST CAISI agent-hijacking evaluations (Jan 2025) https://www.nist.gov/news-events/news/2025/01/technical-blog-strengthening-ai-agent-hijacking-evaluations; Claude Code auto mode figures via Help Net Security (10 Aug 2026) https://www.helpnetsecurity.com/2026/08/10/anthropic-claude-code-auto-mode/ [vendor figures]; OpenAI Codex Auto-review (30 Apr 2026) https://alignment.openai.com/auto-review [vendor]; Microsoft Agent 365 GA (1 May 2026) https://www.microsoft.com/security/blog/2026/05/01/microsoft-agent-365-now-generally-available-expands-capabilities-and-integrations/ [vendor]; Entra Agent ID release notes https://learn.microsoft.com/en-us/entra/agent-id/whats-new-agent-id [vendor]; Okta early access and GA schedule (16 Mar 2026) https://www.okta.com/blog/ai/okta-ai-agents-early-access-announcement/ [vendor]; SailPoint GA via IT Brief Australia (4 Aug 2026) https://itbrief.com.au/story/sailpoint-launches-unified-identity-security-for-ai-agents [secondary]; CyberArk, now Idira, product blog (Jan 2026) https://www.cyberark.com/resources/product-insights-blog/cyberark-secure-ai-agents [vendor]; Ping Identity press release (27 May 2026) https://press.pingidentity.com/2026-05-27-Ping-Identity-Redefines-the-Identity-Control-Plane-for-the-Agentic-Enterprise [vendor]; IETF OAuth working group documents https://datatracker.ietf.org/wg/oauth/documents/; NIST control overlays for securing AI systems https://csrc.nist.gov/projects/cosais; ASD ISM OSCAL releases https://github.com/AustralianCyberSecurityCentre/ism-oscal/releases, with control numbers from a compliance firm's walkthrough of the September 2026 update https://www.cipherprojects.com/blog/posts/asd-ism-september-2026-update/ [secondary]; APRA letter to industry on AI (30 Apr 2026) https://www.apra.gov.au/apra-letter-to-industry-on-artificial-intelligence-ai; MAS media release on AI risk management guidelines (7 Oct 2026) https://www.mas.gov.sg/news/media-releases/2026/mas-sets-out-supervisory-expectations-on-responsible-ai-adoption-by-financial-institutions; Deloitte Australia refund figure: Australian Greens media release (9 Oct 2025) https://greens.org.au/news/media-release/greens-slam-deloittes-unethical-behaviour-and-measly-refund-over-ai-report and CFO Dive (21 Oct 2025) https://www.cfodive.com/news/deloitte-refunds-60k-report-ai-errors-australian-government-accounting/803321/.