@sreetej510/pi-shipd-checks 0.4.4 → 0.4.6

package/README.md

@@ -51,9 +51,10 @@ repeated requests sequentially after applying each result.
 
 - **Solver**: reviewer model and thinking level for the final solver-result reviewer, plus the
   solver model, thinking level, timeout, parallel solver count, and artifact-saving setting.
-- **Analyze Tool**: separate models + thinking levels for the agent-callable gap-analysis and Test
-  Audit modes. The audit model defaults to the gap-analysis model until explicitly changed. These
-  are stored under `analyzeGap.testAuditProvider`, `analyzeGap.testAuditModelId`, and
+- **Analyze Tool**: timeout plus separate models + thinking levels for the agent-callable gap-analysis
+  and Test Audit modes. The timeout applies to each read-only agent phase. The audit model defaults
+  to the gap-analysis model until explicitly changed. These are stored under
+  `analyzeGap.timeoutMinutes`, `analyzeGap.testAuditProvider`, `analyzeGap.testAuditModelId`, and
   `analyzeGap.testAuditThinkingLevel`. Fargate resources are selected automatically and are not
   configured in this menu.
 
@@ -124,8 +125,8 @@ On-Demand fallback. Spot interruptions are retried according to `fargate.maxRetr
 
    `cluster`, `subnetIds`, and `securityGroupId` are optional when a default VPC is available.
    With `adaptiveResourceProfile: true`, the first run uses `resourceProfile`; later runs upgrade
-   when normalized CPU is at least 90% for more than three minutes, downgrade when CPU is at
-   least 95% for under one minute, and otherwise retain the profile. Only the selected next
+   when normalized CPU is at least 95% for more than seven minutes, downgrade when CPU is at
+   least 95% for under two minutes, and otherwise retain the profile. Only the selected next
    profile is written to `projectProfiles`; CPU telemetry is included in `shipd_report.json` but
    no telemetry history is retained. Set `projectProfiles` to override resources per repository:
    `{"C:/path/to/repo":"large"}`.