System Prompt Boundary Bypass
Exploit weak boundaries between system and user prompts to override system-level instructions
Initial Access · Stage 2
Gain control over agent behavior through prompt manipulation or input exploitation
SOUL-INJECT
Directly manipulating or overriding the agent's system-level instructions and behavioral boundaries
Reproduced in a controlled lab environment (DVAA) with documented steps.
L3-04
Reproductions in Damn Vulnerable AI Agent, the OpenA2A intentionally-broken agent for kill-chain validation.
Recent observations
Technique-attributed observations from the AgentPwn + TrapMyAgent fleet, classified into Threat Matrix techniques by the registry evidence bridge. See the Attack Prevalence Index for the full distribution.
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.
Delimiter-escape tiers exploit weak system/user prompt boundaries.
Evidence by source
Evidence timeline
Client fetched the canary embedded in APWN-PI-004 on agentpwn.com (prompt-injection tier 4; classified browser/claimed)
Client fetched the canary embedded in APWN-PI-004 on agentpwn.com (prompt-injection tier 4; classified browser/claimed)
Client fetched the canary embedded in APWN-PI-004 on agentpwn.com (prompt-injection tier 4; classified browser/claimed)
Client fetched the canary embedded in APWN-PI-004 on agentpwn.com (prompt-injection tier 4; classified browser/claimed)
Client fetched the canary embedded in APWN-PI-004 on agentpwn.com (prompt-injection tier 4; classified browser/claimed)
Client fetched the canary embedded in APWN-PI-004 on agentpwn.com (prompt-injection tier 4; classified browser/claimed)
Client fetched the canary embedded in APWN-PI-004 on agentpwn.com (prompt-injection tier 4; classified browser/claimed)
Client fetched the canary embedded in APWN-PI-004 on agentpwn.com (prompt-injection tier 4; classified llm_crawler/claimed)
Client fetched the canary embedded in APWN-PI-004 on agentpwn.com (prompt-injection tier 4; classified browser/claimed)
Client fetched the canary embedded in APWN-PI-004 on agentpwn.com (prompt-injection tier 4; classified llm_crawler/claimed)
Client fetched the canary embedded in APWN-PI-004 on agentpwn.com (prompt-injection tier 4; classified browser/claimed)
Client fetched the canary embedded in APWN-PI-004 on agentpwn.com (prompt-injection tier 4; classified browser/claimed)
Client fetched the canary embedded in APWN-PI-004 on agentpwn.com (prompt-injection tier 4; classified browser/claimed)
Client fetched the canary embedded in APWN-PI-004 on agentpwn.com (prompt-injection tier 4; classified browser/claimed)
Client fetched the canary embedded in APWN-PI-004 on agentpwn.com (prompt-injection tier 4; classified browser/claimed)
Client fetched the canary embedded in APWN-PI-004 on agentpwn.com (prompt-injection tier 4; classified browser/claimed)
Client fetched the canary embedded in APWN-PI-004 on agentpwn.com (prompt-injection tier 4; classified browser/claimed)
Client fetched the canary embedded in APWN-PI-004 on agentpwn.com (prompt-injection tier 4; classified llm_crawler/verified)
Client fetched the canary embedded in APWN-PI-004 on agentpwn.com (prompt-injection tier 4; classified browser/claimed)
Client fetched the canary embedded in APWN-PI-004 on agentpwn.com (prompt-injection tier 4; classified browser/claimed)
Client fetched the canary embedded in APWN-PI-004 on agentpwn.com (prompt-injection tier 4; classified http_client/claimed)
Client fetched the canary embedded in APWN-PI-004 on agentpwn.com (prompt-injection tier 4; classified browser/claimed)
Client fetched the canary embedded in APWN-PI-004 on agentpwn.com (prompt-injection tier 4; classified browser/claimed)
Client fetched the canary embedded in APWN-PI-004 on agentpwn.com (prompt-injection tier 4; classified browser/claimed)
Client fetched the canary embedded in APWN-PI-004 on agentpwn.com (prompt-injection tier 4; classified browser/claimed)
Detection · HackMyAgent
npx hackmyagent secure --ciLive = implemented in hackmyagent; queued = declaredDefense · OASB controls
How to cite
AI Agent Threat Matrix T-2008 (System Prompt Boundary Bypass). OpenA2A, 2026. https://threats.opena2a.org/techniques/T-2008