TL;DR: Indirect prompt injection becomes dangerous when malicious instructions are hidden in pages, tickets, files, and tool outputs that agents already trust, according to KonaSense. Agent controls have to move to action-time authorization and blocking when untrusted content can reach tools and writes.
Editorial analysis by NHI Mgmt Group, based on content published by KonaSense: “Prompt Injection Is a Supply-Chain Problem Now”.
Key questions
Q: Why do indirect prompt injections create more risk than direct jailbreak attempts?
A: Indirect prompt injection is riskier because the agent treats the malicious text as ordinary working context and may act on it through tools, writes, or outbound messages.
Q: How should security teams govern AI agents that call APIs instead of using a UI?
A: Security teams should govern AI agents by treating each callable action as a scoped entitlement, not as a general application login.
Q: What breaks when an AI agent can turn untrusted content into privileged actions?
A: The trust boundary breaks because the agent no longer just reads external text.
Practitioner guidance
- Enforce input provenance checks Classify every page, ticket, file, and tool response the agent consumes as untrusted unless it comes from a known controlled source.
- Separate reasoning from execution Require a policy decision before any command, API call, file write, or outbound connection.
- Instrument detection at the action boundary Monitor both inbound prompts and outbound tool intents for instruction manipulation patterns, then stop the action before the tool runs.
Bottom line: Indirect prompt injection is dangerous because the attacker hides instructions in the content an agent already trusts.
Glossary terms: Indirect Prompt Injection, Action Boundary, Task-Context Poisoning, Agent Runtime
Explore further
View Full Forum → | NHI Foundation Course → | Our Services → | Read the full analysis →
Indirect prompt injection is the core agentic AI governance gap. The risk is not that an AI system says something wrong, but that it treats attacker-authored content as authoritative instructions. That collapses the boundary between untrusted data and permitted action. In practice, organisations need to treat input provenance as a governance control, not a content-filtering nicety, because the same prompt channel is now carrying both information and command.
A few things that frame the scale:
- 19% of organisations give AI systems dramatically more access than human employees, nearly one in five granting unrestricted privilege, according to the 2026 Infrastructure Identity Survey.
- 70% of organisations grant AI systems more access than they would give a human employee performing the exact same job, according to the 2026 Infrastructure Identity Survey.
A question worth separating out:
Q: What is the difference between prompt injection and data poisoning in LLM security?
A: Prompt injection is a runtime attack that manipulates live model inputs to override intended behavior or extract information. Data poisoning happens earlier, when training or fine-tuning data is corrupted to embed bias, backdoors, or unsafe behavior. Teams need different controls for each: input filtering and policy enforcement for injection, and data provenance, review, and validation for poisoning.
👉 Read our full editorial: Indirect prompt injection turns agent inputs into a supply-chain risk