pi-model-fallback 0.5.0 → 0.5.1

package/CHANGELOG.md

@@ -2,6 +2,12 @@
 
 ## Unreleased
 
+## [0.5.1] - 2026-09-30
+
+### Changed
+
+- Update `@earendil-works/pi-*` dependencies to `0.99.1`.
+
 ## [0.5.0] - 2026-09-22
 
 ### Added

package/README.md

@@ -97,7 +97,7 @@ Rule fields:
 Top-level fields:
 
 - `enabled`: toggle the extension without deleting rules
-- `autoRetry`: when a fallback fires from a failed turn, automatically re-queue the failed user prompt so the turn retries on the fallback model (defaults to `true`; set to `false` to only switch models). The failed prompt is queued once, not retried indefinitely.
+- `autoRetry`: when a fallback fires from a failed turn, automatically re-queue the failed user prompt so the turn retries on the fallback model (defaults to `true`; set to `false` to only switch models). The failed prompt is queued once per fallback transition, so a fallback chain may replay it more than once.
 
 Rules use first-match order: the first rule whose provider/model and status match wins. Put specific `matchModels` rules before broad `matchProviders` rules when they should take priority. `model_fallback_config validate`, `save`, `read`, and `status` report warning-only diagnostics when a later rule or model entry is completely shadowed by an earlier rule; `/model-fallback:status` also includes a concise warning summary for the current config. Warning-bearing config remains valid and can still be saved.
 
@@ -124,7 +124,7 @@ If the package is installed project-locally and the current project references i
 - Successful responses do nothing.
 - Matching failures from `after_provider_response` can trigger fallback immediately.
 - Assistant error messages parsed at `turn_end` can also persist fallback state for SDK/provider failures that do not emit the normal response hook. Status extraction looks for HTTP-style tokens (for example `status 429`, `HTTP 503`, or `rate limit`) rather than any bare 3-digit number in the message.
-- With the default `autoRetry: true`, the failed prompt is automatically queued once on the fallback model. Set `autoRetry: false` to switch models without replaying it.
+- With the default `autoRetry: true`, the failed prompt is automatically queued once per fallback transition on the fallback model. A fallback chain may therefore replay the prompt more than once. Set `autoRetry: false` to switch models without replaying it.
 
 Rules never fall back to the failing model itself; a rule whose fallback target equals the failing model is skipped so the next matching rule (or no rule) applies.
 

package/docs/usage.md

@@ -6,7 +6,7 @@ Install:
 pi install npm:pi-model-fallback
 ```
 
-Default behavior: when a `zai/*` model receives `429`, `500`, `502`, `503`, or `504`, Pi switches to `deepseek/deepseek-v4-flash` for the next prompt. The failed prompt is automatically queued once on the fallback model; set `"autoRetry": false` in `model-fallback/config.json` to switch models without retrying it.
+Default behavior: when a `zai/*` model receives `429`, `500`, `502`, `503`, or `504`, Pi switches to `deepseek/deepseek-v4-flash` for the next prompt. The failed prompt is automatically queued once per fallback transition on the fallback model, so a fallback chain may replay it more than once; set `"autoRetry": false` in `model-fallback/config.json` to switch models without retrying it.
 
 Commands:
 

package/lib/error-status.ts

@@ -15,7 +15,30 @@ export function parseReasonFromErrorMessage(message: string): string | undefined
   return undefined;
 }
 
+// Failure modes where the transport reports HTTP success but the provider
+// yields nothing usable — no numeric status is present in any of them, so the
+// status parser must synthesize one for the fallback rules to fire:
+// - "Provider returned an empty response" (OpenRouter stealth/pool models)
+// - "Stream ended without finish_reason" (chunk stream truncated mid-flight)
+// - "<provider> response has no body" (Anthropic/Mistral/generic, 200 empty body)
+// - "<provider> stream ended without a terminal event" (SSE stream dies silently)
+const EMPTY_PROVIDER_RESPONSE_PATTERNS: RegExp[] = [
+  /provider returned an empty response/i,
+  /response (has|with) no body/i,
+  /stream ended without a terminal event/i,
+  /stream ended without finish_reason/i,
+];
+
 export function parseStatusFromErrorMessage(message: string): number | undefined {
+  for (const pattern of EMPTY_PROVIDER_RESPONSE_PATTERNS) {
+    if (pattern.test(message)) {
+      // Synthesize 502 so status-based fallback rules fire instead of the
+      // agent silently retrying the same model. 502 is deliberately chosen
+      // over a novel code: it lands in the 5xx status lists users already
+      // configure and inherits the standard 10-minute cooldown.
+      return 502;
+    }
+  }
   const patterns: RegExp[] = [
     // Leading bare status code, optionally prefixed with "Error:", e.g. "401: {\"type\":\"CreditsError\"...}"
     // or "Error: 401: {\"type\":\"CreditsError\"...}" (OpenCode Go, OpenRouter relays)

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "pi-model-fallback",
-  "version": "0.5.0",
+  "version": "0.5.1",
   "description": "Pi extension that switches to a fallback model after provider failures such as 429 rate limits.",
   "type": "module",
   "license": "MIT",
@@ -50,11 +50,11 @@
     "typebox": "*"
   },
   "devDependencies": {
-    "@earendil-works/pi-agent-core": "^0.86.1",
-    "@earendil-works/pi-ai": "^0.86.1",
-    "@earendil-works/pi-coding-agent": "^0.86.1",
-    "@earendil-works/pi-tui": "^0.86.1",
-    "@types/node": "^26.0.0",
+    "@earendil-works/pi-agent-core": "^0.99.1",
+    "@earendil-works/pi-ai": "^0.99.1",
+    "@earendil-works/pi-coding-agent": "^0.99.1",
+    "@earendil-works/pi-tui": "^0.99.1",
+    "@types/node": "^26.6.3",
     "jiti": "^2.7.0",
     "typebox": "latest",
     "typescript": "^7.0.2"