# Mission drift

When a non-deterministic agent gradually diverges from its intended task. Tracked at runtime so it can be contained before consequences land.

Learn more: [AI Observability](/content/learn/ai-observability/index.html)

Part of [the Agent Control Fabric](/content/platform/index.html): Highflame's identity, policy, and enforcement substrate for AI agents.

## Keep exploring the glossary.

[**Non-human identity (NHI)**  
Identities belonging to machines, services, and agents rather than people. Agents are the fastest-growing and least-governed class of NHI.  
Read →](/content/glossary/nhi/index.html)  
[**On-behalf-of (OBO) chain**  
The unbroken provenance recorded on a credential (who authorized the action, what scope was granted, and how deep the delegation goes) so an audit walks back to a human.  
Read →](/content/glossary/obo-chain/index.html)  
[**Prompt injection**  
An attack that manipulates an agent through crafted input (in a prompt, a tool result, or retrieved content) to make it act against policy.  
Read →](/content/glossary/prompt-injection/index.html)  
[**Red teaming**  
Continuous adversarial testing of AI systems (jailbreaks, extraction, manipulation) with findings turned into enforcement policy and re-scanned to prove the fix.  
Read →](/content/glossary/red-teaming/index.html)  
[**RFC 8693 (token exchange)**  
The standard that lets one token be exchanged for another with attenuated scope: the basis for verifiable agent-to-agent delegation.  
Read →](/content/glossary/rfc-8693/index.html)  
[**Scope attenuation**  
Narrowing permissions at each delegation hop so a sub-agent can never hold more authority than the agent that delegated to it.  
Read →](/content/glossary/scope-attenuation/index.html)