Issue
Action Changed Something Else Too
An AI or agent action makes the requested change but also changes another object, field, file, state, rule, or workflow element that was not supposed to change.
What This Looks Like
The AI or agent performs the requested action, but another thing changes too. A field, file, section, setting, record, rule, or downstream state may be altered even though the user only asked for one bounded change. The requested result may be present, but it comes with an unintended side effect.
Why It Matters
Unintended side effects make AI-assisted work harder to trust. The user cannot review only the requested change; they must inspect surrounding state to see what else moved. In workflows with tools or agents, a small action can become risky if it quietly propagates beyond its declared target.
Structural Signal
An action has effects outside its declared change surface. The issue is not simply that the action failed; it is that the action succeeded in one area while also producing an undeclared secondary effect.
Common Triggers
- The allowed change surface is not explicitly bounded
- The agent applies a fix globally instead of locally
- A tool operation has broader effects than the prompt describes
- The AI rewrites context around the target change
- A repair step updates dependent objects without declaring them
- The workflow does not compare intended and actual change sets
When to Use This Issue
Use this Issue when an AI or agent action makes the requested change but also changes something else that was not supposed to be affected.
When Not to Use This Issue
Do not use this Issue when the unrelated change was explicitly requested or required. Do not use it when the main issue is only that the requested change failed. This Issue applies when there is an unintended secondary effect.
Category
Primary Pattern
Declared Patterns
Undeclared Side Effect
A structural condition where an operation modifies regions outside its declared target scope without a declared propagation or side-effect rule.
Propagation Amplification
A structural condition where an effect, declaration, authority, constraint, or state increases in scope or intensity as it propagates beyond declared bounds.
Boundary Leakage
A structural condition where effects, permissions, state, data, or authority cross a declared boundary without an authorized exception.
Derived Primary Lenses
Escalation Gradient Lens
Measures structural growth or intensification across sequential executions, states, or transitions.
Isolation Boundary Lens
Evaluates whether structural constraints, effects, and regions remain contained within declared boundaries.
Overreach Lens
Detects structure that exceeds declared scope or authority reach.
Propagation Lens
Traces how structural declarations, effects, or state changes propagate across boundaries or stages.
Derived Secondary Lenses
Related AI-Adjacent Issues
Generated Output Changed Hidden Metadata
Generated output changes hidden metadata, identifiers, references, state, or system fields that affect later behavior but are not obvious in the visible output.
Tool Call Caused Hidden Downstream Change
A tool call causes a downstream state, workflow, metadata, record, or integration change that is not visible at the point of action.
Tool Effect Crosses Target Boundary
A tool, connector, function, or integration affects an object, file, record, workspace, or state surface beyond the intended target.
Search Intents
- action changed something else too
- AI changed unrelated thing too
- agent side effect
- fix changed other file
- AI action had unintended side effect
- requested change affected something else