Parser Differential Exploitation
Exploit differences in how parsers (JSON, YAML, markdown) interpret the same input to bypass security controls
Initial Access · Stage 2
Gain control over agent behavior through prompt manipulation or input exploitation
PARSER-DIFFERENTIAL
Exploits differences between parser implementations to bypass security controls
Reproduced in a controlled lab environment (DVAA) with documented steps.
parser-differential-json scenario validates JSON comment and duplicate key attacks
Reproductions in Damn Vulnerable AI Agent, the OpenA2A intentionally-broken agent for kill-chain validation.
AgentPwn coverage
An AgentPwn trap page produces a payload tagged with this technique class. Following the AgentPwn taxonomy of trap pages shows what an agent encounters.
JSON-LD, markdown, and meta-tag delivery tiers exercise parser-differential injection.
Detection · HackMyAgent
npx hackmyagent secure --ciLive = implemented in hackmyagent; queued = declaredDefense · OASB controls
How to cite
AI Agent Threat Matrix T-2009 (Parser Differential Exploitation). OpenA2A, 2026. https://threats.opena2a.org/techniques/T-2009