Guardrails

Workspace

Decision history

Review every recorded shadow recommendation and active guardrail action in this workspace.

All decisions

2075 total decisions, newest first.

When	Rule	Action	Requested → Served	Mode	Saved
Aug 6, 08:17 PM	Reuse exact helpdesk responses	cache exact	gpt-5.5 → cache exact	active	$5
Aug 6, 08:15 PM	Reuse semantically equivalent knowledge answers	cache semantic	gpt-5.5 → cache semantic	active	$6
Aug 6, 08:13 PM	Cap repeated knowledge retrieval	cap retries	gpt-5.5 → gpt-5.5	active	$4
Aug 6, 08:11 PM	Block requests beyond tenant policy	deny	gpt-5.5 → policy	active	$6
Aug 6, 07:46 PM	Use Haiku for validated ticket summaries	swap	claude-sonnet-4-6 → claude-haiku-4-5	active	$12
Aug 6, 07:44 PM	Downgrade routine quality reviews	downgrade	gpt-5.5 → gpt-4o	active	$10
Aug 6, 07:42 PM	Route low-risk chat requests	route	gpt-5.5 → gpt-4o	active	$8
Aug 6, 07:40 PM	Reuse exact helpdesk responses	cache exact	gpt-5.5 → cache exact	active	$5
Aug 6, 07:38 PM	Reuse semantically equivalent knowledge answers	cache semantic	gpt-5.5 → cache semantic	active	$6
Aug 6, 07:36 PM	Cap repeated knowledge retrieval	cap retries	gpt-5.5 → gpt-5.5	active	$4
Aug 6, 07:34 PM	Block requests beyond tenant policy	deny	gpt-5.5 → policy	active	$6
Aug 6, 07:09 PM	Use Haiku for validated ticket summaries	swap	claude-sonnet-4-6 → claude-haiku-4-5	active	$12
Aug 6, 07:07 PM	Downgrade routine quality reviews	downgrade	gpt-5.5 → gpt-4o	active	$10
Aug 6, 07:05 PM	Route low-risk chat requests	route	gpt-5.5 → gpt-4o	active	$8
Aug 6, 07:03 PM	Reuse exact helpdesk responses	cache exact	gpt-5.5 → cache exact	active	$5
Aug 6, 07:01 PM	Reuse semantically equivalent knowledge answers	cache semantic	gpt-5.5 → cache semantic	active	$6
Aug 6, 06:59 PM	Cap repeated knowledge retrieval	cap retries	gpt-5.5 → gpt-5.5	active	$4
Aug 6, 06:57 PM	Block requests beyond tenant policy	deny	gpt-5.5 → policy	active	$6
Aug 6, 06:32 PM	Use Haiku for validated ticket summaries	swap	claude-sonnet-4-6 → claude-haiku-4-5	active	$12
Aug 6, 06:30 PM	Downgrade routine quality reviews	downgrade	gpt-5.5 → gpt-4o	active	$10
Aug 6, 06:28 PM	Route low-risk chat requests	route	gpt-5.5 → gpt-4o	active	$8
Aug 6, 06:26 PM	Reuse exact helpdesk responses	cache exact	gpt-5.5 → cache exact	active	$5
Aug 6, 06:24 PM	Reuse semantically equivalent knowledge answers	cache semantic	gpt-5.5 → cache semantic	active	$6
Aug 6, 06:22 PM	Cap repeated knowledge retrieval	cap retries	gpt-5.5 → gpt-5.5	active	$4
Aug 6, 06:20 PM	Block requests beyond tenant policy	deny	gpt-5.5 → policy	active	$6