PENGO Workbench
Workbench / Lenses / Isolation Boundary Lens

LEN-0220

Isolation Boundary Lens

Evaluates whether structural constraints, effects, and regions remain contained within declared boundaries.

Primary Pattern Matches

PAT-0140

Boundary Leakage

A structural condition where effects, permissions, state, data, or authority cross a declared boundary without an authorized exception.

Lens Application

The Isolation Boundary Lens is a strong inspection mechanism for Boundary Leakage because it focuses attention on whether declared isolation regions actually constrain movement of effects, permissions, state, data, and authority. It helps inspect whether any cross-boundary behavior is explicitly authorized.

Inspect For

  • Effects escaping declared regions
  • Shared state or data crossing boundaries
  • Permission or authority bleed-through
  • Missing or implicit boundary exceptions

Avoid

Treat cross-boundary behavior as an inspection signal, not proof that Boundary Leakage is present.

PAT-0260

Cross-Layer Escalation

A structural condition where escalation originating in one layer propagates into another authority, boundary, or control layer without explicit authorization.

Lens Application

The Isolation Boundary Lens is a strong inspection mechanism for Cross-Layer Escalation because it focuses on whether escalation paths remain contained within intended isolation boundaries. It helps surface cases where actions, privileges, side effects, or control influence cross from one layer into another without a declared handoff, constraint, or authorization point.

Inspect For

  • Escalation paths crossing declared layer boundaries
  • Control effects reaching unintended authority regions
  • Missing authorization at boundary transitions
  • Shared mechanisms that bypass isolation constraints

Avoid

Treat boundary crossing as a signal for closer inspection, not proof that Cross-Layer Escalation is present.

PAT-0400

Unbounded Scope

A structural condition where an operation, authority, rule, or expansion has no declared upper bound, boundary, or termination condition.

Lens Application

The Isolation Boundary Lens is a primary inspection mechanism for Unbounded Scope because it focuses attention on whether the scoped operation remains contained within stated regions, roles, systems, or timeframes. It helps surface places where boundaries are missing, permeable, recursive, or overridden in ways that allow expansion beyond intended containment.

Inspect For

  • Missing or vague scope limits
  • Effects crossing declared regions or roles
  • Expansion paths without termination checks
  • Boundary exceptions that become normal behavior

Avoid

Treat containment failures as inspection signals rather than confirmation that Unbounded Scope is present.

Secondary Pattern Matches

PAT-0420

Undeclared Side Effect

A structural condition where an operation modifies regions outside its declared target scope without a declared propagation or side-effect rule.

Lens Application

As a secondary lens, the Isolation Boundary Lens can support inspection of Undeclared Side Effects by focusing attention on containment assumptions around target scopes, affected regions, and declared effect boundaries. It may surface side-effect signals when observed modifications appear outside the operation’s stated isolation limits.

Inspect For

  • Changes in regions outside the declared target scope
  • Missing propagation or side-effect declarations
  • Boundary crossings not reflected in operation metadata
  • Shared state, caches, or linked regions affected indirectly

Avoid

Treat containment failures as inspection signals requiring further analysis, not as proof that an Undeclared Side Effect has occurred.

Related Issues

ISS-0046

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.

ISS-0113

Agent Modified Unrelated State

An agent changes state outside the requested task, target, file, record, workflow, or authorized scope.

ISS-0009

AI Adds Work Not Requested

The AI adds tasks, steps, analysis, checks, changes, or follow-up work that the user did not ask for.

ISS-0109

AI Memory Has No Governance

Saved or persistent AI memory affects output without clear rules for ownership, scope, review, update, expiry, or removal.

ISS-0114

AI Memory Updated Without Asking

AI memory, saved context, preference, or durable state is updated without the user clearly asking for or approving that update.

ISS-0045

AI Touches Unrelated Scope

The AI affects, edits, analyzes, changes, or reasons over material outside the scope of the requested task.

ISS-0044

AI Uses External Information When Forbidden

The AI uses outside knowledge, sources, tools, memory, or assumptions after the task forbids external information or limits the allowed source set.

ISS-0010

Answer Has Too Many Paths

The answer presents too many possible paths, interpretations, options, or next steps without enough structure to choose among them.

ISS-0111

Automation Skips Required Approval

An automated AI or workflow step proceeds past an approval gate that should have been required before action.

ISS-0032

Context Leaks Between Tasks

Context, assumptions, constraints, examples, files, or decisions from one task affect another task where they should not apply.

ISS-0048

File-Bounded Task Uses Outside Content

The AI is asked to work only from a specific file or document but uses content, assumptions, or sources outside that file.

ISS-0023

Hallucinated Fields

The AI adds fields, keys, attributes, columns, or structured elements that were not declared, requested, or allowed by the expected schema.

ISS-0039

Hidden Rule Overrides Visible Instruction

A hidden, upstream, system, policy, tool, or product rule changes or overrides the visible instruction the user expects the AI to follow.

ISS-0101

Local Exception Grows Into Policy

A local exception, special case, or one-off allowance begins to function like a general policy.

ISS-0096

Local Rule Spreads to Broader Cases

A rule intended for one local case, file, context, user, workflow, or exception begins affecting broader cases.

ISS-0050

Model Output Triggers Unapproved Action

AI output causes, recommends, or triggers an action that has not passed the required approval, permission, or authority check.

ISS-0012

No Owner for Agent Action

An agent action can affect the system without a declared responsible owner, authority, or accountable decision path.

ISS-0049

Policy Exception Spreads Too Far

A narrow policy exception, allowance, or special case spreads beyond its intended scope and begins governing broader cases.

ISS-0076

Prompt Does Not Say What to Exclude

The prompt declares what to include but does not declare what should be excluded, allowing unwanted scope, sources, content, or actions into the result.

ISS-0047

Repair Step Created New Breakage

A repair, correction, retry, or fix step addresses one problem but introduces a new failure elsewhere.

ISS-0062

Retrieval Exceeds Evidence Limit

The AI retrieves, uses, cites, or considers more evidence than the task permits or more than the review surface can support.

ISS-0051

Review Outcome Changes Unrelated Environment

A review result, approval, rejection, or classification changes state outside the environment, case, file, or workflow it was meant to govern.

ISS-0112

Routing Overrides Task Intent

Routing, mode selection, agent behavior, or workflow classification sends the task down a path that overrides what the user was trying to accomplish.

ISS-0103

Single Step Carries Too Many Decisions

One prompt, workflow step, review stage, or agent action carries too many decisions for the system or user to evaluate cleanly.

ISS-0093

Small Error Spreads Into Large Failure

A small AI, output, routing, or workflow error propagates through later steps until it becomes a larger failure.

ISS-0011

Task Has No Clear Limit

The task does not declare where the AI should stop, what is out of scope, or what counts as enough work.

ISS-0061

Too Many Tool Calls

The AI or agent makes more tool calls, searches, retrievals, API calls, or integration actions than the task requires or permits.

ISS-0108

Tool Can Act Without Responsible Authority

A tool, connector, function, or integration can perform an action without a declared responsible authority for that action.