OWASP
OWASP Foundation · Open Web Application Security Project · OWASP GenAI Security Project
Open Web Application Security Project: the nonprofit that maintains the OWASP Top 10 list of web-application risks and the OWASP Top 10 for Large Language Model Applications.
In our briefings
The claim each of the newest briefings on OWASP would have you leave with.
OWASP's AIVSS v0.8 starts from a CVSS v4.0 base score and adds an uplift worth part of the gap between that base and 10. Ten agentic factors, each scored 0, 0.5 or 1, set the share, and an exploit-maturity multiplier scales it, by default 0.97. With the default mitigation factor of 1.0 (no or weak mitigation), all ten worked examples in the document land between 7.1 and 10.0, seven Critical and three High, and a goal-manipulation flaw with a CVSS base of 2.1 comes out at 7.1. The authors read a jump like that as a sign that the agentic deployment context, not the technical flaw, is the main risk driver. On our reading of the formula, a score can only fall below its CVSS base through the mitigation factor, and the document warns that AIVSS scores are not directly comparable to CVSS base scores.
OWASP's AIVSS lifts a CVSS 2.1 agent flaw to 7.1, and all ten worked examples land High or CriticalBI-FIND-2026-2299
Under this draft a client's identity is a web address and whatever its owner publishes there. The client_id is an https URL, which revision -03 says "MUST contain a path component other than /", and the JSON document at that URL "MUST contain a client_id property whose value MUST match" it. The authorization server learns the client's name, redirect addresses and keys by fetching that document. Its security section says a client that changes its URL is "a brand new client with no relationship to the previous one", and that losing control of the domain would let a third party "assume the client's identity". We read this as a name anyone can look up. That is more than an anonymous registration gives a server, though it says nothing about what software sits behind the name.
OAuth's URL client ID draft tightens server rules in -03; the client is still whoever controls the URLBI-FIND-2026-0319
Frederick of Mozilla says the team's 2025 attempts at having LLMs read code produced long reports, "sometimes hallucinating", and Mozilla went no further until it built a harness. Once software can tell "whether there is truly a bug based on hard deterministic facts, you don't have false positives anymore," he says. Dan of Cloudflare says what a harness needs: mechanical verification, reproducing steps "that you know exist", file and code references, and separate stages of mechanical and LLM-based checks. "You can't just hand an LLM code base and say, find a bug," he says. Neither puts the result down to one model: Frederick does not think good vulnerability research needs the best model, and Steve Springett calls it a misconception that much of what Mythos does cannot be done today.
Mozilla's AI bug hunting got past hallucinated reports once a harness checked each finding mechanicallyBI-FIND-2026-1507
›Cited in 26 briefings
Every briefing that discusses this term, newest first.
- OWASP's AIVSS lifts a CVSS 2.1 agent flaw to 7.1, and all ten worked examples land High or Critical
Article · · OWASP AIVSS project · Ken Huang et al. · Oct 22, 2026
- OAuth's URL client ID draft tightens server rules in -03; the client is still whoever controls the URL
Draft · · IETF Datatracker · Aaron Parecki et al. · Jul 6, 2026
- Mozilla's AI bug hunting got past hallucinated reports once a harness checked each finding mechanically
Video · · OWASP Foundation · Uploaded October 21, 2026
- Five places to put authorization in microservices, and why Drus says the classic gateway one does not scale
Video · · OWASP Foundation · Uploaded October 21, 2026
- Agents and MCP servers get broad standing access because it is quick or easy, three voices say, and three vendors now pitch a gateway in front of MCP
Field signals · · 9 voices from the field · Sep 2025 to Oct 2026
- The OWASP MCP Top 10 lead says few security teams can find the MCP servers developers run on their own machines
Podcast · · The Application Security Podcast · Shadow MCP Servers: The AI Security Risk Hiding on Your Developers' Laptops · Oct 12, 2026
- At OWASP, one poisoned memory entry made a demo agent send a Q4 report to an attacker server
Video · · OWASP Foundation · Recorded Jun 2026 · Uploaded October 21, 2026
- From Level 2 up, OWASP's AI security standard asks for an identity per agent and a check on every MCP tool call
Video · · OWASP Foundation · Oct 21, 2026
- A quarter of 31,132 agent skills matched a risky pattern; 23 of the 87 riskiest showed clear signs of malice
Paper · · arXiv · Yi Liu et al. · Jan 15, 2026
- Anthropic asks for per-agent identity; its guide leaves Claude Code sub-agent credentials unspecified
Article · · Anthropic (Claude) · Anthropic · May 27, 2026
- Anthropic says agents should never use static API keys, yet its sub-agents can carry the parent's permissions
Article · · Anthropic · Anthropic · May 18, 2026
- OWASP's agent Top 10 blames identity systems built for people, and does not say how it chose or ranked its risks
Article · · OWASP GenAI Security Project · John Sotiropoulos et al. · Dec 9, 2025
- Corvic's own tests find its agent vault keeps keys out of prompts but does much less against a misused connector
Paper · · arXiv · Kenney, Corvic AI · Sep 27, 2026
- OpenID agent-claims draft adds a provider-set user, application or agent label; scores stay each provider's own
Draft · · IETF Datatracker · Raza Sharif, CyberSecAI · Mar 26, 2026
- A cross-domain agent trust draft requires the call chain, then leaves its format to a version that never came
Draft · · IETF Datatracker · Daniel King et al. · Mar 15, 2026
- An MCP signing draft shows a client who wrote a tool, and admits a poisoned tool can still be validly signed
Draft · · IETF Datatracker · Raza Sharif, CyberSecAI · Mar 14, 2026
- MACAW Security's trust layer signs every agent call, but what stopped its prompt-injection test was a policy
Paper · · arXiv · Mohan Rajagopalan et al. · Feb 11, 2026
- The draft replacing RFC 8725 requires rejecting malformed JWTs but only recommends denial-of-service limits
Draft · · IETF datatracker · Yaron Sheffer et al. · Jan 12, 2026
- AAGATE maps NIST's AI risk framework onto Kubernetes, and its paper never times the 'millisecond kill switch'
Paper · · arXiv · Ken Huang et al. · Oct 29, 2025
- Agentic JWT binds an agent's token to its task and its own code, but it is one author's draft, not a standard
Paper · · arXiv (cs.CR preprint 2509.13597) · Abhishek Goswami · Sep 16, 2025
- A 'provable' bound on agent-memory injection assumes independent layers and fills in none of its detection rates
Paper · · arXiv · Ken Huang et al. · Aug 17, 2025
- A code scanner's top find in 583 MCP servers was hard-coded keys; tool poisoning needed an MCP-specific scanner
Paper · · arXiv · Mohammed Mehedi Hasan et al. · Jun 16, 2025
- The first Agent Name Service paper puts capabilities in the name and leaves protocol translation and governance out
Paper · · arXiv (cs.CR preprint 2505.10609) / OWASP Agentic Security Initiative · Ken Huang et al. · May 15, 2025
- OWASP's multi-agent threat guide finds the hop where a user's permission check drops out, but names no fix
Article · · OWASP GenAI Security Project · OWASP Agentic Security Initiative · Apr 22, 2025
- A fake error message relayed between agents got three multi-agent frameworks to run an attacker's code
Paper · · arXiv · Harold Triedman et al. · Mar 15, 2025
- MAESTRO names your AI agents' identity threats and leaves designing the controls to you
Article · · Cloud Security Alliance · Ken Huang · Feb 6, 2025
Related terms
What appears alongside OWASP in the same briefings.
OWASP Agent Control Standard (ACS)OWASP Top 10 for LLM Applications 2025