dsh-shift-router

Configuration

Configuration lives in the shift-router settings namespace: edit it in the GUI (Settings → shift-router), with /router config commands, or via the profile patch row. All fields have safe defaults.

Field	Default	Description
enabled	true	Master switch
tiers.fast.models	[]	Fast-tier chain (provider/model + priority); also the Judge's model chain
tiers.smart.models	[]	Smart-tier chain
routing.mode	auto	auto (default): judge + routing + failover + orchestration; manual: no judge, only explicit /route-force overrides; off: fully passive for model selection (commands/telemetry still work)
routing.judgeTimeout	5000	Judge call timeout (ms)
routing.judgeMaxTokens	4000	Max output tokens for a single Judge call
routing.judgePromptCap	6000	Max prompt characters sent to the Judge (bounds Judge cost)
routing.window.size	5	Downgrade sliding window size
routing.window.threshold	0.6	Fast-majority ratio required to downgrade
routing.window.minConfidence	0.5	Ignore judge verdicts below this confidence
routing.cacheAware.enabled	true	Same-provider cache protection
routing.cacheAware.sameFamilyThreshold	0.9	Downgrade threshold when tiers share a provider
routing.cacheAware.idleBoundaryMs	300000	Idle gap before a warm cache is considered cold
orchestration.mode	auto	auto: complex → Smart CTO; off: plain two-tier routing
orchestration.maxRounds	3	Delegate→review rounds hard cap (enforced: each subagent delegation counts one round; at the cap the subagent tool is denied)
orchestration.escalationThreshold	2	Failed worker results before Smart must take over the phase (enforced: each isError subagent result counts)
orchestration.requireSmartModel	true	Skip orchestration if the Smart model can't be resolved
failover.baseMs	60000	Cooldown base delay for 5xx failures (1m)
failover.maxMs	21600000	Hard cap on the backoff ladder (6h)
failover.startAttempts4xx	3	4xx (429/quota) failures start at this attempt (16m), client limits usually outlive server blips
failover.speedWindowSize	5	Recent tokens/sec readings kept for the /router stats average
telemetry.callLogCap	1000	Max per-message attribution records kept for baseline cost computation
ux.routerLogVerbose	false	Print router decisions to the harness log
pricing	[]	Optional {provider, model, input, output, cacheRead?, cacheWrite?} USD-per-1M-token table for cost telemetry

All numeric fields are range-validated by the schema (e.g. window.threshold must be in [0,1], window.size a positive integer); invalid values are rejected at load / on set, never silently accepted.