gsd-pi 2.44.0 → 2.45.0-dev.1afbdaa

package/src/resources/extensions/gsd/auto/infra-errors.ts

@@ -18,6 +18,9 @@ export const INFRA_ERROR_CODES: ReadonlySet<string> = new Set([
   "EDQUOT",   // disk quota exceeded
   "EMFILE",   // too many open files (process)
   "ENFILE",   // too many open files (system)
+  "ECONNREFUSED", // connection refused (offline / local server down)
+  "ENOTFOUND",    // DNS lookup failed (offline / no network)
+  "ENETUNREACH",  // network unreachable (offline / no route)
 ]);
 
 /**

package/src/resources/extensions/gsd/auto/loop-deps.ts

@@ -80,7 +80,6 @@ export interface LoopDeps {
     basePath: string,
     unitType: string,
     unitId: string,
-    completedUnits: number,
     sessionFile?: string,
   ) => void;
   handleLostSessionLock: (
@@ -179,29 +178,11 @@ export interface LoopDeps {
     startedAt: number,
     opts?: CloseoutOptions & Record<string, unknown>,
   ) => Promise<void>;
-  verifyExpectedArtifact: (
-    unitType: string,
-    unitId: string,
-    basePath: string,
-  ) => boolean;
-  clearUnitRuntimeRecord: (
-    basePath: string,
-    unitType: string,
-    unitId: string,
-  ) => void;
-  writeUnitRuntimeRecord: (
-    basePath: string,
-    unitType: string,
-    unitId: string,
-    startedAt: number,
-    record: Record<string, unknown>,
-  ) => void;
   recordOutcome: (unitType: string, tier: string, success: boolean) => void;
   writeLock: (
     lockBase: string,
     unitType: string,
     unitId: string,
-    completedCount: number,
     sessionFile?: string,
   ) => void;
   captureAvailableSkills: () => void;

package/src/resources/extensions/gsd/auto/phases.ts

@@ -24,10 +24,15 @@ import {
 import { detectStuck } from "./detect-stuck.js";
 import { runUnit } from "./run-unit.js";
 import { debugLog } from "../debug-logger.js";
-import { gsdRoot } from "../paths.js";
-import { atomicWriteSync } from "../atomic-write.js";
 import { PROJECT_FILES } from "../detection.js";
+import { MergeConflictError } from "../git-service.js";
 import { join } from "node:path";
+import { existsSync, cpSync } from "node:fs";
+import { logWarning, logError } from "../workflow-logger.js";
+import { gsdRoot } from "../paths.js";
+import { atomicWriteSync } from "../atomic-write.js";
+import { verifyExpectedArtifact } from "../auto-recovery.js";
+import { writeUnitRuntimeRecord } from "../unit-runtime.js";
 
 // ─── generateMilestoneReport ──────────────────────────────────────────────────
 
@@ -162,8 +167,8 @@ export async function runPreDispatch(
       debugLog("autoLoop", { phase: "exit", reason: "health-gate-failed" });
       return { action: "break", reason: "health-gate-failed" };
     }
-  } catch {
-    // Non-fatal
+  } catch (e) {
+    logWarning("engine", "Pre-dispatch health gate threw unexpectedly", { error: String(e) });
   }
 
   // Sync project root artifacts into worktree
@@ -233,26 +238,24 @@ export async function runPreDispatch(
     loopState.stuckRecoveryAttempts = 0;
 
     // Worktree lifecycle on milestone transition — merge current, enter next
-    deps.resolver.mergeAndExit(s.currentMilestoneId!, ctx.ui);
-
-    // Opt-in: create draft PR on milestone completion
-    if (prefs?.git?.auto_pr) {
-      try {
-        const { createDraftPR } = await import("../git-service.js");
-        const prUrl = createDraftPR(
-          s.basePath,
-          s.currentMilestoneId!,
-          `[GSD] ${s.currentMilestoneId} complete`,
-          `Milestone ${s.currentMilestoneId} completed by GSD auto-mode.\n\nSee .gsd/${s.currentMilestoneId}/ for details.`,
+    try {
+      deps.resolver.mergeAndExit(s.currentMilestoneId!, ctx.ui);
+    } catch (mergeErr) {
+      if (mergeErr instanceof MergeConflictError) {
+        // Real code conflicts — stop the loop instead of retrying forever (#2330)
+        ctx.ui.notify(
+          `Merge conflict: ${mergeErr.conflictedFiles.join(", ")}. Resolve conflicts manually and run /gsd auto to resume.`,
+          "error",
         );
-        if (prUrl) {
-          ctx.ui.notify(`Draft PR created: ${prUrl}`, "info");
-        }
-      } catch {
-        // Non-fatal — PR creation is best-effort
+        await deps.stopAuto(ctx, pi, `Merge conflict on milestone ${s.currentMilestoneId}`);
+        return { action: "break", reason: "merge-conflict" };
       }
+      // Non-conflict merge errors — log and continue
+      logWarning("engine", "Milestone merge failed with non-conflict error", { milestone: s.currentMilestoneId!, error: String(mergeErr) });
     }
 
+    // PR creation (auto_pr) is handled inside mergeMilestoneToMain (#2302)
+
     deps.invalidateAllCaches();
 
     state = await deps.deriveState(s.basePath);
@@ -276,14 +279,20 @@ export async function runPreDispatch(
       .map((m: { id: string }) => m.id);
     deps.pruneQueueOrder(s.basePath, pendingIds);
 
-    // Reset completed-units tracking for the new milestone — stale entries
-    // from the previous milestone cause the dispatch loop to skip units
-    // that haven't actually been completed in the new milestone's context.
-    s.completedUnits = [];
+    // Archive the old completed-units.json instead of wiping it (#2313).
     try {
       const completedKeysPath = join(gsdRoot(s.basePath), "completed-units.json");
+      if (existsSync(completedKeysPath) && s.currentMilestoneId) {
+        const archivePath = join(
+          gsdRoot(s.basePath),
+          `completed-units-${s.currentMilestoneId}.json`,
+        );
+        cpSync(completedKeysPath, archivePath);
+      }
       atomicWriteSync(completedKeysPath, JSON.stringify([], null, 2));
-    } catch { /* non-fatal */ }
+    } catch (e) {
+      logWarning("engine", "Failed to archive completed-units on milestone transition", { error: String(e) });
+    }
 
     // Rebuild STATE.md immediately so it reflects the new active milestone.
     // This bypasses the 30-second throttle in the normal rebuild path —
@@ -291,8 +300,8 @@ export async function runPreDispatch(
     // immediate write.
     try {
       await deps.rebuildState(s.basePath);
-    } catch {
-      // Non-fatal — STATE.md will be rebuilt on the next regular cycle
+    } catch (e) {
+      logWarning("engine", "STATE.md rebuild failed after milestone transition", { error: String(e) });
     }
   }
 
@@ -322,25 +331,20 @@ export async function runPreDispatch(
     if (incomplete.length === 0 && state.registry.length > 0) {
       // All milestones complete — merge milestone branch before stopping
       if (s.currentMilestoneId) {
-        deps.resolver.mergeAndExit(s.currentMilestoneId, ctx.ui);
-
-        // Opt-in: create draft PR on milestone completion
-        if (prefs?.git?.auto_pr) {
-          try {
-            const { createDraftPR } = await import("../git-service.js");
-            const prUrl = createDraftPR(
-              s.basePath,
-              s.currentMilestoneId,
-              `[GSD] ${s.currentMilestoneId} complete`,
-              `Milestone ${s.currentMilestoneId} completed by GSD auto-mode.\n\nSee .gsd/${s.currentMilestoneId}/ for details.`,
+        try {
+          deps.resolver.mergeAndExit(s.currentMilestoneId, ctx.ui);
+        } catch (mergeErr) {
+          if (mergeErr instanceof MergeConflictError) {
+            ctx.ui.notify(
+              `Merge conflict: ${mergeErr.conflictedFiles.join(", ")}. Resolve conflicts manually and run /gsd auto to resume.`,
+              "error",
             );
-            if (prUrl) {
-              ctx.ui.notify(`Draft PR created: ${prUrl}`, "info");
-            }
-          } catch {
-            // Non-fatal — PR creation is best-effort
+            await deps.stopAuto(ctx, pi, `Merge conflict on milestone ${s.currentMilestoneId}`);
+            return { action: "break", reason: "merge-conflict" };
           }
         }
+
+        // PR creation (auto_pr) is handled inside mergeMilestoneToMain (#2302)
       }
       deps.sendDesktopNotification(
         "GSD",
@@ -422,25 +426,20 @@ export async function runPreDispatch(
   if (state.phase === "complete") {
     // Milestone merge on complete (before closeout so branch state is clean)
     if (s.currentMilestoneId) {
-      deps.resolver.mergeAndExit(s.currentMilestoneId, ctx.ui);
-
-      // Opt-in: create draft PR on milestone completion
-      if (prefs?.git?.auto_pr) {
-        try {
-          const { createDraftPR } = await import("../git-service.js");
-          const prUrl = createDraftPR(
-            s.basePath,
-            s.currentMilestoneId,
-            `[GSD] ${s.currentMilestoneId} complete`,
-            `Milestone ${s.currentMilestoneId} completed by GSD auto-mode.\n\nSee .gsd/${s.currentMilestoneId}/ for details.`,
+      try {
+        deps.resolver.mergeAndExit(s.currentMilestoneId, ctx.ui);
+      } catch (mergeErr) {
+        if (mergeErr instanceof MergeConflictError) {
+          ctx.ui.notify(
+            `Merge conflict: ${mergeErr.conflictedFiles.join(", ")}. Resolve conflicts manually and run /gsd auto to resume.`,
+            "error",
           );
-          if (prUrl) {
-            ctx.ui.notify(`Draft PR created: ${prUrl}`, "info");
-          }
-        } catch {
-          // Non-fatal — PR creation is best-effort
+          await deps.stopAuto(ctx, pi, `Merge conflict on milestone ${s.currentMilestoneId}`);
+          return { action: "break", reason: "merge-conflict" };
         }
       }
+
+      // PR creation (auto_pr) is handled inside mergeMilestoneToMain (#2302)
     }
     deps.sendDesktopNotification(
       "GSD",
@@ -539,7 +538,7 @@ export async function runDispatch(
       if (loopState.stuckRecoveryAttempts === 0) {
         // Level 1: try verifying the artifact, then cache invalidation + retry
         loopState.stuckRecoveryAttempts++;
-        const artifactExists = deps.verifyExpectedArtifact(
+        const artifactExists = verifyExpectedArtifact(
           unitType,
           unitId,
           s.basePath,
@@ -848,7 +847,7 @@ export async function runUnitPhase(
   const unitStartSeq = ic.nextSeq();
   deps.emitJournalEvent({ ts: new Date().toISOString(), flowId: ic.flowId, seq: unitStartSeq, eventType: "unit-start", data: { unitType, unitId } });
   deps.captureAvailableSkills();
-  deps.writeUnitRuntimeRecord(
+  writeUnitRuntimeRecord(
     s.basePath,
     unitType,
     unitId,
@@ -860,6 +859,7 @@ export async function runUnitPhase(
       lastProgressAt: s.currentUnit.startedAt,
       progressCount: 0,
       lastProgressKind: "dispatch",
+      recoveryAttempts: 0, // Reset so re-dispatched units get full recovery budget (#2322)
     },
   );
 
@@ -921,8 +921,8 @@ export async function runUnitPhase(
         (decisionsContent?.length ?? 0) +
         (requirementsContent?.length ?? 0) +
         (projectContent?.length ?? 0);
-    } catch {
-      // Non-fatal
+    } catch (e) {
+      logWarning("engine", "Baseline char count measurement failed", { error: String(e) });
     }
   }
 
@@ -932,9 +932,7 @@ export async function runUnitPhase(
   } catch (reorderErr) {
     const msg =
       reorderErr instanceof Error ? reorderErr.message : String(reorderErr);
-    process.stderr.write(
-      `[gsd] prompt reorder failed (non-fatal): ${msg}\n`,
-    );
+    logWarning("engine", "Prompt reorder failed", { error: msg });
   }
 
   // Select and apply model (with tier escalation on retry — normal units only)
@@ -1001,7 +999,6 @@ export async function runUnitPhase(
     deps.lockBase(),
     unitType,
     unitId,
-    s.completedUnits.length,
   );
 
   debugLog("autoLoop", {
@@ -1032,14 +1029,12 @@ export async function runUnitPhase(
     deps.lockBase(),
     unitType,
     unitId,
-    s.completedUnits.length,
     sessionFile,
   );
   deps.writeLock(
     deps.lockBase(),
     unitType,
     unitId,
-    s.completedUnits.length,
     sessionFile,
   );
 
@@ -1103,8 +1098,8 @@ export async function runUnitPhase(
           `${unitType} ${unitId} completed with 0 tool calls — hallucinated summary, will retry`,
           "warning",
         );
-        // Do NOT add to completedUnits — fall through to next iteration
-        // where dispatch will re-derive and re-dispatch this task.
+        // Fall through to next iteration where dispatch will re-derive
+        // and re-dispatch this task.
         return { action: "next", data: { unitStartedAt: s.currentUnit.startedAt } };
       }
     }
@@ -1121,25 +1116,8 @@ export async function runUnitPhase(
   const skipArtifactVerification = unitType.startsWith("hook/") || unitType === "custom-step";
   const artifactVerified =
     skipArtifactVerification ||
-    deps.verifyExpectedArtifact(unitType, unitId, s.basePath);
+    verifyExpectedArtifact(unitType, unitId, s.basePath);
   if (artifactVerified) {
-    s.completedUnits.push({
-      type: unitType,
-      id: unitId,
-      startedAt: s.currentUnit.startedAt,
-      finishedAt: Date.now(),
-    });
-    if (s.completedUnits.length > 200) {
-      s.completedUnits = s.completedUnits.slice(-200);
-    }
-    // Flush completed-units to disk so the record survives crashes
-    try {
-      const completedKeysPath = join(gsdRoot(s.basePath), "completed-units.json");
-      const keys = s.completedUnits.map((u) => `${u.type}/${u.id}`);
-      atomicWriteSync(completedKeysPath, JSON.stringify(keys, null, 2));
-    } catch { /* non-fatal: disk flush failure */ }
-
-    deps.clearUnitRuntimeRecord(s.basePath, unitType, unitId);
     s.unitDispatchCount.delete(`${unitType}/${unitId}`);
     s.unitRecoveryCount.delete(`${unitType}/${unitId}`);
   }
@@ -1184,8 +1162,8 @@ export async function runFinalize(
   // Sidecar items use lightweight pre-verification opts
   const preVerificationOpts: PreVerificationOpts | undefined = sidecarItem
     ? sidecarItem.kind === "hook"
-      ? { skipSettleDelay: true, skipDoctor: true, skipStateRebuild: true, skipWorktreeSync: true }
-      : { skipSettleDelay: true, skipStateRebuild: true }
+      ? { skipSettleDelay: true, skipWorktreeSync: true }
+      : { skipSettleDelay: true }
     : undefined;
   const preResult = await deps.postUnitPreVerification(postUnitCtx, preVerificationOpts);
   if (preResult === "dispatched") {

package/src/resources/extensions/gsd/auto/run-unit.ts

@@ -11,6 +11,7 @@ import { NEW_SESSION_TIMEOUT_MS } from "./session.js";
 import type { UnitResult } from "./types.js";
 import { _setCurrentResolve, _setSessionSwitchInFlight } from "./resolve.js";
 import { debugLog } from "../debug-logger.js";
+import { logWarning, logError } from "../workflow-logger.js";
 
 /**
  * Execute a single unit: create a new session, send the prompt, and await
@@ -85,7 +86,9 @@ export async function runUnit(
     if (process.cwd() !== s.basePath) {
       process.chdir(s.basePath);
     }
-  } catch { /* non-fatal — chdir may fail if dir was removed */ }
+  } catch (e) {
+    logWarning("engine", "Failed to chdir to basePath before dispatch", { basePath: s.basePath, error: String(e) });
+  }
 
   // ── Send the prompt ──
   debugLog("runUnit", { phase: "send-message", unitType, unitId });
@@ -115,8 +118,8 @@ export async function runUnit(
     if (typeof cmdCtxAny?.clearQueue === "function") {
       (cmdCtxAny.clearQueue as () => unknown)();
     }
-  } catch {
-    // Non-fatal — clearQueue may not be available in all contexts
+  } catch (e) {
+    logWarning("engine", "clearQueue failed after unit completion", { error: String(e) });
   }
 
   return result;

package/src/resources/extensions/gsd/auto/session.ts

@@ -23,13 +23,6 @@ import type { BudgetAlertLevel } from "../auto-budget.js";
 
 // ─── Exported Types ──────────────────────────────────────────────────────────
 
-export interface CompletedUnit {
-  type: string;
-  id: string;
-  startedAt: number;
-  finishedAt: number;
-}
-
 export interface CurrentUnit {
   type: string;
   id: string;
@@ -106,7 +99,6 @@ export class AutoSession {
   // ── Current unit ─────────────────────────────────────────────────────────
   currentUnit: CurrentUnit | null = null;
   currentUnitRouting: UnitRouting | null = null;
-  completedUnits: CompletedUnit[] = [];
   currentMilestoneId: string | null = null;
 
   // ── Model state ──────────────────────────────────────────────────────────
@@ -160,14 +152,6 @@ export class AutoSession {
     return this.originalBasePath || this.basePath;
   }
 
-  completeCurrentUnit(): CompletedUnit | null {
-    if (!this.currentUnit) return null;
-    const done: CompletedUnit = { ...this.currentUnit, finishedAt: Date.now() };
-    this.completedUnits.push(done);
-    this.currentUnit = null;
-    return done;
-  }
-
   reset(): void {
     this.clearTimers();
 
@@ -193,7 +177,6 @@ export class AutoSession {
     // Unit
     this.currentUnit = null;
     this.currentUnitRouting = null;
-    this.completedUnits = [];
     this.currentMilestoneId = null;
 
     // Model
@@ -234,7 +217,6 @@ export class AutoSession {
       activeRunDir: this.activeRunDir,
       currentMilestoneId: this.currentMilestoneId,
       currentUnit: this.currentUnit,
-      completedUnits: this.completedUnits.length,
       unitDispatchCount: Object.fromEntries(this.unitDispatchCount),
     };
   }

package/src/resources/extensions/gsd/auto-artifact-paths.ts

@@ -0,0 +1,131 @@
+// GSD Auto-mode — Artifact Path Resolution
+//
+// resolveExpectedArtifactPath and diagnoseExpectedArtifact moved here from
+// auto-recovery.ts (Phase 5 dead-code cleanup). The artifact verification
+// function was removed entirely — callers now query WorkflowEngine directly.
+
+import {
+  resolveMilestonePath,
+  resolveSlicePath,
+  relMilestoneFile,
+  relSliceFile,
+  buildMilestoneFileName,
+  buildSliceFileName,
+  buildTaskFileName,
+} from "./paths.js";
+import { join } from "node:path";
+
+/**
+ * Resolve the expected artifact for a unit to an absolute path.
+ */
+export function resolveExpectedArtifactPath(
+  unitType: string,
+  unitId: string,
+  base: string,
+): string | null {
+  const parts = unitId.split("/");
+  const mid = parts[0]!;
+  const sid = parts[1];
+  switch (unitType) {
+    case "discuss-milestone": {
+      const dir = resolveMilestonePath(base, mid);
+      return dir ? join(dir, buildMilestoneFileName(mid, "CONTEXT")) : null;
+    }
+    case "research-milestone": {
+      const dir = resolveMilestonePath(base, mid);
+      return dir ? join(dir, buildMilestoneFileName(mid, "RESEARCH")) : null;
+    }
+    case "plan-milestone": {
+      const dir = resolveMilestonePath(base, mid);
+      return dir ? join(dir, buildMilestoneFileName(mid, "ROADMAP")) : null;
+    }
+    case "research-slice": {
+      const dir = resolveSlicePath(base, mid, sid!);
+      return dir ? join(dir, buildSliceFileName(sid!, "RESEARCH")) : null;
+    }
+    case "plan-slice": {
+      const dir = resolveSlicePath(base, mid, sid!);
+      return dir ? join(dir, buildSliceFileName(sid!, "PLAN")) : null;
+    }
+    case "reassess-roadmap": {
+      const dir = resolveSlicePath(base, mid, sid!);
+      return dir ? join(dir, buildSliceFileName(sid!, "ASSESSMENT")) : null;
+    }
+    case "run-uat": {
+      const dir = resolveSlicePath(base, mid, sid!);
+      return dir ? join(dir, buildSliceFileName(sid!, "UAT-RESULT")) : null;
+    }
+    case "execute-task": {
+      const tid = parts[2];
+      const dir = resolveSlicePath(base, mid, sid!);
+      return dir && tid
+        ? join(dir, "tasks", buildTaskFileName(tid, "SUMMARY"))
+        : null;
+    }
+    case "complete-slice": {
+      const dir = resolveSlicePath(base, mid, sid!);
+      return dir ? join(dir, buildSliceFileName(sid!, "SUMMARY")) : null;
+    }
+    case "validate-milestone": {
+      const dir = resolveMilestonePath(base, mid);
+      return dir ? join(dir, buildMilestoneFileName(mid, "VALIDATION")) : null;
+    }
+    case "complete-milestone": {
+      const dir = resolveMilestonePath(base, mid);
+      return dir ? join(dir, buildMilestoneFileName(mid, "SUMMARY")) : null;
+    }
+    case "replan-slice": {
+      const dir = resolveSlicePath(base, mid, sid!);
+      return dir ? join(dir, buildSliceFileName(sid!, "REPLAN")) : null;
+    }
+    case "rewrite-docs":
+      return null;
+    case "reactive-execute":
+      // Reactive execute produces multiple task summaries — verified separately
+      return null;
+    default:
+      return null;
+  }
+}
+
+export function diagnoseExpectedArtifact(
+  unitType: string,
+  unitId: string,
+  base: string,
+): string | null {
+  const parts = unitId.split("/");
+  const mid = parts[0];
+  const sid = parts[1];
+  switch (unitType) {
+    case "discuss-milestone":
+      return `${relMilestoneFile(base, mid!, "CONTEXT")} (milestone context from discussion)`;
+    case "research-milestone":
+      return `${relMilestoneFile(base, mid!, "RESEARCH")} (milestone research)`;
+    case "plan-milestone":
+      return `${relMilestoneFile(base, mid!, "ROADMAP")} (milestone roadmap)`;
+    case "research-slice":
+      return `${relSliceFile(base, mid!, sid!, "RESEARCH")} (slice research)`;
+    case "plan-slice":
+      return `${relSliceFile(base, mid!, sid!, "PLAN")} (slice plan)`;
+    case "execute-task": {
+      const tid = parts[2];
+      return `Task ${tid} marked [x] in ${relSliceFile(base, mid!, sid!, "PLAN")} + summary written`;
+    }
+    case "complete-slice":
+      return `Slice ${sid} marked [x] in ${relMilestoneFile(base, mid!, "ROADMAP")} + summary + UAT written`;
+    case "replan-slice":
+      return `${relSliceFile(base, mid!, sid!, "REPLAN")} + updated ${relSliceFile(base, mid!, sid!, "PLAN")}`;
+    case "rewrite-docs":
+      return "Active overrides resolved in .gsd/OVERRIDES.md + plan documents updated";
+    case "reassess-roadmap":
+      return `${relSliceFile(base, mid!, sid!, "ASSESSMENT")} (roadmap reassessment)`;
+    case "run-uat":
+      return `${relSliceFile(base, mid!, sid!, "UAT-RESULT")} (UAT result)`;
+    case "validate-milestone":
+      return `${relMilestoneFile(base, mid!, "VALIDATION")} (milestone validation report)`;
+    case "complete-milestone":
+      return `${relMilestoneFile(base, mid!, "SUMMARY")} (milestone summary)`;
+    default:
+      return null;
+  }
+}

package/src/resources/extensions/gsd/auto-dashboard.ts

@@ -48,7 +48,6 @@ export interface AutoDashboardData {
   startTime: number;
   elapsed: number;
   currentUnit: { type: string; id: string; startedAt: number } | null;
-  completedUnits: { type: string; id: string; startedAt: number; finishedAt: number }[];
   basePath: string;
   /** Running cost and token totals from metrics ledger */
   totalCost: number;