@zhixuan92/multi-model-agent 3.4.0 → 3.5.1

package/README.md

@@ -88,6 +88,26 @@ The auth token is generated on first `mmagent serve`. Retrieve it with `mmagent
 
 All tool endpoints require bearer auth: `Authorization: Bearer <token>`.
 
+## What's new in 3.5.1
+
+**Bug fixes:**
+- **Single-provider deployments no longer burn a doomed cross-tier fallback call.** When `agents.standard` and `agents.complex` resolve to the same backend (one-provider deployment) and the assigned-tier call transport-fails, the wrapper used to substitute to the alt tier — which in that configuration just hits the same backend, burning a second doomed call and surfacing as `terminationReason: 'all_tiers_unavailable'`. The original failure now flows through as the task's terminal result with the actual root-cause status. No new operator config; auto-detected via deep-equal of the effective provider config.
+- **No more `runner_crash: verbose-line: invalid key name` on fallback / rework paths.** With `diagnostics.verbose: true`, any run that hit fallback / escalation / spec_rework / quality_rework previously threw inside the verbose-stream serializer (camelCase event-param keys like `assignedTier`, `implTier`, `attemptCap` violated its snake_case-only validator) and surfaced as terminal `runner_crash` even though the model itself succeeded. The verbose-stream branch now drops `batchId` / `taskIndex` (already emitted as `batch` / `task`) and snake-cases the remaining keys; the JSONL `DiagnosticLogger` contract (camelCase `assignedTier` / `implTier` / ... on `escalation` / `fallback` events) is unchanged.
+
+## What's new in 3.5.0
+
+**Breaking changes (operators read this first):**
+- `task.maxReviewRounds` is gone — review caps now derive from policy tables (`maxReworksFor('spec') = 2`, `maxReworksFor('quality') = 2`). Remove the field from any callers.
+- `agentType` is gone from `/execute-plan` (top-level + per-task). The compiler hardcodes `agentType: 'standard'`. `/delegate` is unchanged and still accepts the field.
+- Status-level escalation inside `delegateWithEscalation` is removed. Transport failures now flow through the new `runWithFallback` wrapper in `reviewed-lifecycle.ts`.
+
+**New behavior:**
+- **Tier-escalating rework.** For standard-tier tasks, the implementation tier escalates to complex on the final rework attempt; reviewers swap to keep impl ≠ reviewer.
+- **Runtime tier fallback.** Transport failures (`api_error` / `network_error` / `timeout`) or missing configuration trigger automatic substitution of the other tier. Fallback is sticky per loop.
+- **Single-slot operators** receive reviews on the same tier (`violatesSeparation: true`); set `reviewPolicy: 'off'` to opt out.
+- **Four new diagnostic events** — `escalation`, `escalation_unavailable`, `fallback`, `fallback_unavailable` — emitted via the verbose stderr stream and JSONL log.
+- **New `agents.*History` and `agents.fallbackOverrides`** envelope fields surface tier movement; the headline composer adds `(escalated to complex; fallback fired)` style suffixes.
+
 ## Operator commands
 
 ```bash

package/dist/http/handlers/tools/execute-plan.d.ts.map

@@ -1 +1 @@
-{"version":3,"file":"execute-plan.d.ts","sourceRoot":"","sources":["../../../../src/http/handlers/tools/execute-plan.ts"],"names":[],"mappings":"AAOA,OAAO,KAAK,EAAE,WAAW,EAAE,MAAM,uBAAuB,CAAC;AAEzD,OAAO,KAAK,EAAE,UAAU,EAAE,MAAM,iBAAiB,CAAC;AAElD,wBAAgB,uBAAuB,CAAC,IAAI,EAAE,WAAW,GAAG,UAAU,CAuCrE"}
+{"version":3,"file":"execute-plan.d.ts","sourceRoot":"","sources":["../../../../src/http/handlers/tools/execute-plan.ts"],"names":[],"mappings":"AAOA,OAAO,KAAK,EAAE,WAAW,EAAE,MAAM,uBAAuB,CAAC;AAEzD,OAAO,KAAK,EAAE,UAAU,EAAE,MAAM,iBAAiB,CAAC;AAElD,wBAAgB,uBAAuB,CAAC,IAAI,EAAE,WAAW,GAAG,UAAU,CAsDrE"}

package/dist/http/handlers/tools/execute-plan.js

@@ -7,9 +7,27 @@ export function buildExecutePlanHandler(deps) {
     return async (_req, res, _params, ctx) => {
         const parsed = executePlan.inputSchema.safeParse(ctx.body);
         if (!parsed.success) {
-            sendError(res, 400, 'invalid_request', 'Request body validation failed', {
-                fieldErrors: parsed.error.flatten(),
-            });
+            const fieldErrors = {};
+            for (const issue of parsed.error.issues) {
+                let path = issue.path.join('.');
+                if (path === '' && issue.message.includes('"agentType"')) {
+                    path = 'agentType';
+                }
+                else if (path.startsWith('tasks.') && issue.message === 'Invalid input') {
+                    const task = issue.path.reduce((value, segment) => {
+                        if (value && typeof value === 'object')
+                            return value[segment];
+                        return undefined;
+                    }, ctx.body);
+                    if (task && typeof task === 'object' && 'agentType' in task) {
+                        path = `${path}.agentType`;
+                    }
+                }
+                if (!fieldErrors[path])
+                    fieldErrors[path] = [];
+                fieldErrors[path].push(issue.message);
+            }
+            sendError(res, 400, 'invalid_request', 'Request body validation failed', { fieldErrors });
             return;
         }
         const input = parsed.data;

package/dist/http/handlers/tools/execute-plan.js.map

@@ -1 +1 @@
-{"version":3,"file":"execute-plan.js","sourceRoot":"","sources":["../../../../src/http/handlers/tools/execute-plan.ts"],"names":[],"mappings":"AAGA,OAAO,KAAK,WAAW,MAAM,6DAA6D,CAAC;AAC3F,OAAO,EAAE,kBAAkB,EAAE,MAAM,0DAA0D,CAAC;AAC9F,OAAO,EAAE,SAAS,EAAE,QAAQ,EAAE,MAAM,iBAAiB,CAAC;AACtD,OAAO,EAAE,aAAa,EAAE,MAAM,yBAAyB,CAAC;AAExD,OAAO,EAAE,mBAAmB,EAAE,MAAM,gCAAgC,CAAC;AAGrE,MAAM,UAAU,uBAAuB,CAAC,IAAiB;IACvD,OAAO,KAAK,EAAE,IAAqB,EAAE,GAAmB,EAAE,OAA+B,EAAE,GAAG,EAAE,EAAE;QAChG,MAAM,MAAM,GAAG,WAAW,CAAC,WAAW,CAAC,SAAS,CAAC,GAAG,CAAC,IAAI,CAAC,CAAC;QAC3D,IAAI,CAAC,MAAM,CAAC,OAAO,EAAE,CAAC;YACpB,SAAS,CAAC,GAAG,EAAE,GAAG,EAAE,iBAAiB,EAAE,gCAAgC,EAAE;gBACvE,WAAW,EAAE,MAAM,CAAC,KAAK,CAAC,OAAO,EAAE;aACpC,CAAC,CAAC;YACH,OAAO;QACT,CAAC;QAED,MAAM,KAAK,GAAG,MAAM,CAAC,IAAI,CAAC;QAC1B,MAAM,GAAG,GAAG,GAAG,CAAC,GAAI,CAAC;QAErB,MAAM,aAAa,GAAG,IAAI,CAAC,eAAe,CAAC,cAAc,CAAC,GAAG,CAAC,CAAC;QAC/D,IAAI,CAAC,aAAa,CAAC,EAAE,EAAE,CAAC;YACtB,SAAS,CAAC,GAAG,EAAE,GAAG,EAAE,aAAa,CAAC,KAAK,EAAE,aAAa,CAAC,OAAO,CAAC,CAAC;YAChE,OAAO;QACT,CAAC;QACD,MAAM,EAAE,GAAG,aAAa,CAAC,cAAc,CAAC;QACxC,EAAE,CAAC,cAAc,GAAG,IAAI,CAAC,GAAG,EAAE,CAAC;QAC/B,IAAI,CAAC,eAAe,CAAC,iBAAiB,CAAC,GAAG,CAAC,CAAC;QAE5C,MAAM,QAAQ,GAAG,KAAK,CAAC,eAAe,IAAI,EAAE,CAAC;QAC7C,MAAM,EAAE,OAAO,EAAE,SAAS,EAAE,GAAG,aAAa,CAAC;YAC3C,IAAI,EAAE,cAAc;YACpB,UAAU,EAAE,GAAG;YACf,QAAQ;YACR,aAAa,EAAE,IAAI,CAAC,aAAa;YACjC,cAAc,EAAE,EAAE;YAClB,IAAI;YACJ,QAAQ,EAAE,KAAK,EAAE,YAAY,EAAE,EAAE;gBAC/B,OAAO,kBAAkB,CAAC,YAAY,EAAE,KAAK,CAAC,CAAC;YACjD,CAAC;SACF,CAAC,CAAC;QAEH,MAAM,mBAAmB,CAAC,EAAE,MAAM,EAAE,IAAI,CAAC,MAAM,EAAE,OAAO,EAAE,KAAK,EAAE,IAAI,CAAC,GAAG,IAAI,EAAE,EAAE,MAAM,EAAE,KAAK,EAAE,CAAC,CAAC;QAElG,QAAQ,CAAC,GAAG,EAAE,GAAG,EAAE,EAAE,OAAO,EAAE,SAAS,EAAE,CAAC,CAAC;IAC7C,CAAC,CAAC;AACJ,CAAC"}
+{"version":3,"file":"execute-plan.js","sourceRoot":"","sources":["../../../../src/http/handlers/tools/execute-plan.ts"],"names":[],"mappings":"AAGA,OAAO,KAAK,WAAW,MAAM,6DAA6D,CAAC;AAC3F,OAAO,EAAE,kBAAkB,EAAE,MAAM,0DAA0D,CAAC;AAC9F,OAAO,EAAE,SAAS,EAAE,QAAQ,EAAE,MAAM,iBAAiB,CAAC;AACtD,OAAO,EAAE,aAAa,EAAE,MAAM,yBAAyB,CAAC;AAExD,OAAO,EAAE,mBAAmB,EAAE,MAAM,gCAAgC,CAAC;AAGrE,MAAM,UAAU,uBAAuB,CAAC,IAAiB;IACvD,OAAO,KAAK,EAAE,IAAqB,EAAE,GAAmB,EAAE,OAA+B,EAAE,GAAG,EAAE,EAAE;QAChG,MAAM,MAAM,GAAG,WAAW,CAAC,WAAW,CAAC,SAAS,CAAC,GAAG,CAAC,IAAI,CAAC,CAAC;QAC3D,IAAI,CAAC,MAAM,CAAC,OAAO,EAAE,CAAC;YACpB,MAAM,WAAW,GAA6B,EAAE,CAAC;YACjD,KAAK,MAAM,KAAK,IAAI,MAAM,CAAC,KAAK,CAAC,MAAM,EAAE,CAAC;gBACxC,IAAI,IAAI,GAAG,KAAK,CAAC,IAAI,CAAC,IAAI,CAAC,GAAG,CAAC,CAAC;gBAChC,IAAI,IAAI,KAAK,EAAE,IAAI,KAAK,CAAC,OAAO,CAAC,QAAQ,CAAC,aAAa,CAAC,EAAE,CAAC;oBACzD,IAAI,GAAG,WAAW,CAAC;gBACrB,CAAC;qBAAM,IAAI,IAAI,CAAC,UAAU,CAAC,QAAQ,CAAC,IAAI,KAAK,CAAC,OAAO,KAAK,eAAe,EAAE,CAAC;oBAC1E,MAAM,IAAI,GAAG,KAAK,CAAC,IAAI,CAAC,MAAM,CAAU,CAAC,KAAK,EAAE,OAAO,EAAE,EAAE;wBACzD,IAAI,KAAK,IAAI,OAAO,KAAK,KAAK,QAAQ;4BAAE,OAAQ,KAA0C,CAAC,OAA0B,CAAC,CAAC;wBACvH,OAAO,SAAS,CAAC;oBACnB,CAAC,EAAE,GAAG,CAAC,IAAI,CAAC,CAAC;oBACb,IAAI,IAAI,IAAI,OAAO,IAAI,KAAK,QAAQ,IAAI,WAAW,IAAI,IAAI,EAAE,CAAC;wBAC5D,IAAI,GAAG,GAAG,IAAI,YAAY,CAAC;oBAC7B,CAAC;gBACH,CAAC;gBACD,IAAI,CAAC,WAAW,CAAC,IAAI,CAAC;oBAAE,WAAW,CAAC,IAAI,CAAC,GAAG,EAAE,CAAC;gBAC/C,WAAW,CAAC,IAAI,CAAC,CAAC,IAAI,CAAC,KAAK,CAAC,OAAO,CAAC,CAAC;YACxC,CAAC;YACD,SAAS,CAAC,GAAG,EAAE,GAAG,EAAE,iBAAiB,EAAE,gCAAgC,EAAE,EAAE,WAAW,EAAE,CAAC,CAAC;YAC1F,OAAO;QACT,CAAC;QAED,MAAM,KAAK,GAAG,MAAM,CAAC,IAAI,CAAC;QAC1B,MAAM,GAAG,GAAG,GAAG,CAAC,GAAI,CAAC;QAErB,MAAM,aAAa,GAAG,IAAI,CAAC,eAAe,CAAC,cAAc,CAAC,GAAG,CAAC,CAAC;QAC/D,IAAI,CAAC,aAAa,CAAC,EAAE,EAAE,CAAC;YACtB,SAAS,CAAC,GAAG,EAAE,GAAG,EAAE,aAAa,CAAC,KAAK,EAAE,aAAa,CAAC,OAAO,CAAC,CAAC;YAChE,OAAO;QACT,CAAC;QACD,MAAM,EAAE,GAAG,aAAa,CAAC,cAAc,CAAC;QACxC,EAAE,CAAC,cAAc,GAAG,IAAI,CAAC,GAAG,EAAE,CAAC;QAC/B,IAAI,CAAC,eAAe,CAAC,iBAAiB,CAAC,GAAG,CAAC,CAAC;QAE5C,MAAM,QAAQ,GAAG,KAAK,CAAC,eAAe,IAAI,EAAE,CAAC;QAC7C,MAAM,EAAE,OAAO,EAAE,SAAS,EAAE,GAAG,aAAa,CAAC;YAC3C,IAAI,EAAE,cAAc;YACpB,UAAU,EAAE,GAAG;YACf,QAAQ;YACR,aAAa,EAAE,IAAI,CAAC,aAAa;YACjC,cAAc,EAAE,EAAE;YAClB,IAAI;YACJ,QAAQ,EAAE,KAAK,EAAE,YAAY,EAAE,EAAE;gBAC/B,OAAO,kBAAkB,CAAC,YAAY,EAAE,KAAK,CAAC,CAAC;YACjD,CAAC;SACF,CAAC,CAAC;QAEH,MAAM,mBAAmB,CAAC,EAAE,MAAM,EAAE,IAAI,CAAC,MAAM,EAAE,OAAO,EAAE,KAAK,EAAE,IAAI,CAAC,GAAG,IAAI,EAAE,EAAE,MAAM,EAAE,KAAK,EAAE,CAAC,CAAC;QAElG,QAAQ,CAAC,GAAG,EAAE,GAAG,EAAE,EAAE,OAAO,EAAE,SAAS,EAAE,CAAC,CAAC;IAC7C,CAAC,CAAC;AACJ,CAAC"}

package/dist/skills/mma-audit/SKILL.md

@@ -10,7 +10,7 @@ when_to_use: >-
   mmagent is running. Delegate so each file audits on its own worker; the main
   agent only synthesizes findings. Audit on PROSE/SPEC docs — use mma-review for
   source code.
-version: 3.4.0
+version: 3.5.1
 ---
 
 # mma-audit

package/dist/skills/mma-clarifications/SKILL.md

@@ -9,7 +9,7 @@ when_to_use: >-
   mma-debug / mma-investigate terminal envelope has `proposedInterpretation` as
   a string. Read the proposal, decide whether to accept or correct it, then call
   this skill. The batch resumes immediately after the POST returns.
-version: 3.4.0
+version: 3.5.1
 ---
 
 # mma-clarifications

package/dist/skills/mma-context-blocks/SKILL.md

@@ -10,7 +10,7 @@ when_to_use: >-
   Register once here, then pass the ID via `contextBlockIds` on mma-delegate /
   mma-execute-plan / mma-audit / mma-review / mma-verify / mma-debug /
   mma-investigate. Cheaper and faster than inlining the same content N times.
-version: 3.4.0
+version: 3.5.1
 ---
 
 # mma-context-blocks

package/dist/skills/mma-debug/SKILL.md

@@ -10,7 +10,7 @@ when_to_use: >-
   read files, reproduce, trace — OR a methodology skill
   (superpowers:systematic-debugging) points at the investigation step. Delegate
   the read/reproduce/trace; the main agent stays on the hypothesis and the fix.
-version: 3.4.0
+version: 3.5.1
 ---
 
 # mma-debug

package/dist/skills/mma-delegate/SKILL.md

@@ -11,7 +11,7 @@ when_to_use: >-
   and keep main context free. If a plan file exists → use mma-execute-plan. If
   the task is audit / review / verify / debug / investigate → use the matching
   specialized skill.
-version: 3.4.0
+version: 3.5.1
 ---
 
 # mma-delegate

package/dist/skills/mma-execute-plan/SKILL.md

@@ -10,7 +10,7 @@ when_to_use: >-
   superpowers:subagent-driven-development / superpowers:executing-plans —
   workers are cheaper and don't pollute main context. Task descriptors must
   match plan headings verbatim.
-version: 3.4.0
+version: 3.5.1
 ---
 
 # mma-execute-plan

package/dist/skills/mma-investigate/SKILL.md

@@ -10,7 +10,7 @@ when_to_use: >-
   running. Delegate the read/grep/synthesis to a worker so the main context
   stays on judgment. Codebase only — does not perform web research or
   git-history queries.
-version: 3.4.0
+version: 3.5.1
 ---
 
 # mma-investigate

package/dist/skills/mma-retry/SKILL.md

@@ -10,7 +10,7 @@ when_to_use: >-
   you want to re-try the failed indices only. Prefer this over re-dispatching
   the whole batch or inline-retrying — it's idempotent and preserves the
   original batch's diagnostics.
-version: 3.4.0
+version: 3.5.1
 ---
 
 # mma-retry

package/dist/skills/mma-review/SKILL.md

@@ -10,7 +10,7 @@ when_to_use: >-
   AND mmagent is running. Delegate so each file reviews on its own worker; the
   main agent only decides what to merge. Review on SOURCE CODE — use mma-audit
   for prose specs / configs.
-version: 3.4.0
+version: 3.5.1
 ---
 
 # mma-review

package/dist/skills/mma-verify/SKILL.md

@@ -10,7 +10,7 @@ when_to_use: >-
   against implemented work BEFORE claiming success. Delegate so each checklist
   item gets independent evidence-gathering on a worker. Use this BEFORE saying
   "done" — never after.
-version: 3.4.0
+version: 3.5.1
 ---
 
 # mma-verify

package/dist/skills/multi-model-agent/SKILL.md

@@ -11,7 +11,7 @@ when_to_use: >-
   tasks — AND mmagent is running. Read this once, pick the matching mma-* skill,
   and delegate there. Applies equally whether the user invoked a superpowers
   methodology skill or asked directly.
-version: 3.4.0
+version: 3.5.1
 ---
 
 # multi-model-agent (router)

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@zhixuan92/multi-model-agent",
-  "version": "3.4.0",
+  "version": "3.5.1",
   "type": "module",
   "license": "MIT",
   "description": "Standalone HTTP server for multi-model-agent. Routes tool-invocation work to Claude, Codex, or OpenAI-compatible sub-agents with async-polling REST dispatch and installable skills for Claude Code, Gemini CLI, Codex CLI, and Cursor.",
@@ -52,7 +52,7 @@
   },
   "dependencies": {
     "@asteasolutions/zod-to-openapi": "^8.5.0",
-    "@zhixuan92/multi-model-agent-core": "^3.4.0",
+    "@zhixuan92/multi-model-agent-core": "^3.5.1",
     "gray-matter": "^4.0.3",
     "minimist": "^1.2.8",
     "zod": "^4.0.0"