The Fixed Harness & Mutation Budget
The answer is a clean state-space partition. You split the autonomous loop's entire environment into two mutually exclusive regimes: the Fixed Harness contains everything the agent is mathematically barred from altering, the immutable constraints of the enterprise. The Mutation Budget contains everything the agent is explicitly authorized to alter, bounded by a finite computational allowance.
Fig 1The Fixed Harness / Mutation Budget dichotomy. The outer perimeter is immutable; the inner dashed boundary is the agent's finite exploration sandbox.
The Fixed Harness enforces three types of gates (pre-loop, mid-loop, and post-loop) that the agent cannot circumvent. These gates exist at the infrastructure level, not at the prompt level. You are not asking the agent nicely to respect compliance. You are making it physically impossible for the agent to violate it.
Design Principle
If the agent writes code that attempts to bypass any gate (say, disabling the PII redaction filter to access richer features) the Fixed Harness immediately terminates the execution. Not a warning. Not a retry. A hard stop.
class FixedHarness:
"""Immutable enterprise compliance layer.
The agent cannot modify, disable, or circumvent these gates."""
def pre_loop_gate(self, query, data_request):
"""Enforced BEFORE the agent accesses any data."""
data = self.pii_redactor.scrub(data_request)
if query.requests_individual_records():
raise FixedHarnessViolation(
"SYSTEM HALT: Agent attempted individual-level retrieval. "
"RBAC restricts access to cohort-aggregate views only."
)
return data
def mid_loop_gate(self, mutation, current_state):
"""Enforced DURING each mutation step."""
# Verify the mutation does not alter any harness-protected parameter
if mutation.targets_protected_parameter():
raise FixedHarnessViolation(
"SYSTEM HALT: Agent attempted to modify a Fixed Harness parameter."
)
# Verify data access patterns remain within RBAC boundaries
if mutation.introduces_individual_level_access():
raise FixedHarnessViolation(
"SYSTEM HALT: Mutation would breach aggregate-only data policy."
)
def post_loop_gate(self, model_output, fairness_thresholds):
"""Enforced AFTER the agent produces a model."""
fairness_report = self.fairness_evaluator.assess(model_output)
for metric, value in fairness_report.items():
if value < fairness_thresholds[metric]:
raise FixedHarnessViolation(
f"SYSTEM HALT: Model failed fairness gate. "
f"{metric}={value:.3f} < threshold={fairness_thresholds[metric]:.3f}"
)
return model_output
04 / Bounded Exploration
