Delete never-edited/never-acted/no-ship/no-progress salvage classes

Two peer agent loops have no concept of a stop that bars retry, and neither
treats a tool-using run with no net file diff as a failure. This project had
five: no-ship, no-progress, never-acted, never-edited, and repetition all
hard-blocked identical re-dispatch for the rest of the session. A worker
sharing a directory can issue real edits that a concurrent lane absorbs,
leaving no net diff — the never-edited class read that as "did nothing" and
discarded the worker's real work while also blocking retry.

Deletes HARD_BLOCK_SALVAGES/isHardBlockSalvage and the fingerprint-refusal
logic in brief-dispatch.ts (the ledger now always admits), the requireEdit
branch and the consecutive-identical no-progress check in stop-policy.ts
(with their bookkeeping: DEFAULT_SUBAGENT_REPEAT_LIMIT, ToolCallStreak,
nextToolCallStreak, subAgentNoProgress), and the four now-false "it will be
refused" parent hints. shell-evidence.ts's write-detection existed solely to
feed requireEdit and is now dead; shrunk to reads-only (still feeds the
CritiqueDirector requireEvidence gate). editedPaths bookkeeping in thrash.ts
stays for intervention-log diagnostics only.

repetition keeps its detector (a sibling change owns deleting that) but loses
its hard-block membership here, matching every other salvage class: it ends
the run and reports the loop, but does not refuse a later re-dispatch.

Author: TheGreatAxios

CHANGELOG.md

@@ -11,6 +11,18 @@ matching `## [X.Y.Z]` section (plus install instructions). Do not maintain
 parallel copies under `docs/` or `scripts/notes/`. At cut time: rename
 `## [Unreleased]` to `## [X.Y.Z] - YYYY-MM-DD`, then run the release script.
 
+## [Unreleased]
+
+### Agent
+
+- Removed the `never-edited`, `never-acted`, `no-ship`, and `no-progress` leaf
+  salvage classes and the sticky hard-block that refused an identical
+  re-dispatch after any of them fired. A worker sharing a directory can issue
+  real edits that a concurrent writer absorbs, leaving no net diff — that is
+  not a failure, and no salvage class now treats it as one. `turn-budget`,
+  `deadline`, `stalled`, `cancelled`, `incomplete-report`, and `repetition`
+  are unaffected.
+
 ## [0.2.108] - 2026-08-24
 
 ### Agent

src/agent/director.test.ts

@@ -840,13 +840,13 @@ describe("ChatDirector tool-only loop protection", () => {
     expect(actions.some((a) => a.type === "infer")).toBe(true);
   });
 
-  test("after a hard-block salvage, Skywalker is nudged once and unique reads do not pause", async () => {
+  test("after a repetition salvage, Skywalker is nudged once and unique reads do not pause", async () => {
     const director = createChatDirector("system", [], {
       onTasksChange: () => {},
       provider: providerlessPolicy,
     });
     const capabilities = makeCapabilities();
-    const salvage = forcedStopReport("no-ship", "mapped the tree, never edited");
+    const salvage = forcedStopReport("repetition", "looped mid-stream");
     await director.decide(
       {
         type: "inference.done",
@@ -1071,7 +1071,7 @@ describe("ChatDirector tool-only loop protection", () => {
         await director.decide(taskTurn(id), mockState, capabilities);
         const result = actionsArray(
           await director.decide(
-            taskDoneEvent(id, { content: forcedStopReport("no-progress", "x") }),
+            taskDoneEvent(id, { content: forcedStopReport("turn-budget", "x") }),
             mockState,
             capabilities,
           ),

src/agent/director.ts

@@ -31,7 +31,7 @@ import {
 } from "../subagent/stop-policy.js";
 import { PRESENT_VIEW_PRIMITIVES_GUIDANCE } from "./tool-schema-normalize.js";
 import { isOperatorOriginated } from "./message-provenance.js";
-import { classifyBriefSalvage, isHardBlockSalvage } from "../subagent/brief-dispatch.js";
+import { classifyBriefSalvage } from "../subagent/brief-dispatch.js";
 import { PRIMARY_SALVAGE_NUDGE } from "./look-tour.js";
 
 // Fired when turnsSinceUserMessage reaches TURNS_SINCE_USER_MESSAGE_BACKSTOP.
@@ -935,7 +935,7 @@ class ChatDirectorImpl extends DefaultDirector {
       this.pendingTaskCallIds.delete(event.result.callId);
       const body = typeof event.result.content === "string" ? event.result.content : "";
       const salvage = classifyBriefSalvage(body);
-      if (salvage !== null && isHardBlockSalvage(salvage) && !this.salvageNudgeFired) {
+      if (salvage === "repetition" && !this.salvageNudgeFired) {
         this.salvageNudgeFired = true;
         this.pendingSalvageNudge = PRIMARY_SALVAGE_NUDGE;
       }

src/agent/prompts.ts

@@ -153,7 +153,7 @@ export function buildGuidelines(
           "- Prefer the typed spawn contract on every worker: `intent`, `success_criteria` (done-when), `do_not` (scope fence), and `report_focus` so workers finish instead of thrashing. Free-form `prompt` alone is weaker.",
           "- After workers return, merge their Summary/Findings into a coherent answer for the operator; do not paste raw sub-agent dumps.",
           "- Pass `maxTurns` on `task` when a job needs a larger inference budget (default 30, no hard upper cap). On turn-budget salvage, re-dispatch with continuation context and a higher maxTurns only a few times on the same brief — after the re-dispatch cap, change approach instead of bumping turns again.",
-          "- After thrash / no-progress / repetition / never-acted salvage, do not re-dispatch an identical brief (prompt/agent/intent/success_criteria/do_not) — it is refused. Change the brief to force a re-run; maxTurns alone does not unlock it.",
+          "- After a repetition salvage, re-dispatching an identical brief (prompt/agent/intent/success_criteria/do_not) unchanged will likely loop again — change the brief before retrying.",
           "- Use manage_tasks for your own coordination checklist; spawning workers is `task`, not manage_tasks.",
           "- If context is compacted automatically, do not stop tasks early due to token fear; persist progress via manage_tasks and worker reports.",
         ]),

src/subagent/brief-dispatch.ts

@@ -1,12 +1,11 @@
 /**
- * Parent-side re-dispatch caps for task briefs (CL-4343 + CL-5203).
+ * Parent-side re-dispatch tracking for task briefs (CL-4343 + CL-5203).
  *
- * Leaf stops already salvage no-progress / turn-budget / etc. This
- * module tracks how often the *parent* re-spawns the same brief so:
- * - hard-block-class salvages refuse an identical re-dispatch for the rest of
- *   the parent chat session (sticky until the fingerprint changes)
- * - turn-budget salvage flips from "raise maxTurns" to "stop" after enough
- *   same-brief dispatches without a successful complete
+ * Leaf stops already salvage turn-budget / deadline / etc. This module
+ * tracks how often the *parent* re-spawns the same brief so turn-budget
+ * salvage flips from "raise maxTurns" to "stop" after enough same-brief
+ * dispatches without a successful complete. No salvage class refuses
+ * re-dispatch (CL-6994) — every dispatch is admitted.
  *
  * Session-scoped: one ledger per createTaskTool instance (parent chat tool).
  */
@@ -15,20 +14,12 @@ import type { TaskIntent } from "./report.js";
 import {
   isDeadlineSubAgentReport,
   isForcedStopSubAgentReport,
-  isNeverActedSubAgentReport,
-  isNeverEditedSubAgentReport,
-  isNoProgressSubAgentReport,
-  isNoShipSubAgentReport,
   isRepetitionSubAgentReport,
   isTurnBudgetSubAgentReport,
 } from "./stop-policy.js";
 
-/** Salvage classes that must not be re-dispatched with an identical brief. */
-export type HardBlockSalvage =
-  "no-ship" | "no-progress" | "repetition" | "never-acted" | "never-edited";
-
 export type BriefSalvageKind =
-  HardBlockSalvage | "turn-budget" | "deadline" | "stalled" | "cancelled" | "incomplete-report";
+  "turn-budget" | "deadline" | "stalled" | "cancelled" | "incomplete-report" | "repetition";
 
 export interface TaskBriefFingerprintInput {
   prompt: string;
@@ -41,8 +32,6 @@ export interface TaskBriefFingerprintInput {
 export interface BriefDispatchRecord {
   /** How many times this fingerprint has been accepted for run (including first). */
   dispatchCount: number;
-  /** Last salvage class observed for this fingerprint, if any. */
-  lastSalvage?: BriefSalvageKind;
 }
 
 /**
@@ -52,18 +41,6 @@ export interface BriefDispatchRecord {
  */
 export const TURN_BUDGET_STOP_AFTER_DISPATCHES = 3;
 
-const HARD_BLOCK_SALVAGES = new Set<BriefSalvageKind>([
-  "no-ship",
-  "no-progress",
-  "repetition",
-  "never-acted",
-  "never-edited",
-]);
-
-export function isHardBlockSalvage(kind: BriefSalvageKind): kind is HardBlockSalvage {
-  return HARD_BLOCK_SALVAGES.has(kind);
-}
-
 /** True when the worker returned a stall salvage report. */
 export function isStalledSubAgentReport(report: string): boolean {
   return isForcedStopSubAgentReport(report, "stalled");
@@ -85,11 +62,7 @@ export function isIncompleteReportSubAgentReport(report: string): boolean {
  */
 export function classifyBriefSalvage(report: string): BriefSalvageKind | null {
   // Order: more specific salvage phrases first.
-  if (isNoShipSubAgentReport(report)) return "no-ship";
   if (isRepetitionSubAgentReport(report)) return "repetition";
-  if (isNeverEditedSubAgentReport(report)) return "never-edited";
-  if (isNeverActedSubAgentReport(report)) return "never-acted";
-  if (isNoProgressSubAgentReport(report)) return "no-progress";
   if (isTurnBudgetSubAgentReport(report)) return "turn-budget";
   if (isDeadlineSubAgentReport(report)) return "deadline";
   if (isStalledSubAgentReport(report)) return "stalled";
@@ -127,13 +100,8 @@ function serializeList(items: readonly string[] | undefined): string {
 
 export interface BriefDispatchLedger {
   get: (fingerprint: string) => BriefDispatchRecord | undefined;
-  /**
-   * Pre-run gate. Returns ok with the 1-based dispatch count that will be used,
-   * or a reject message for the parent tool result.
-   */
-  admit: (
-    fingerprint: string,
-  ) => { ok: true; dispatchCount: number } | { ok: false; message: string };
+  /** Pre-run gate. Always admits; returns the 1-based dispatch count that will be used. */
+  admit: (fingerprint: string) => { ok: true; dispatchCount: number };
   /** Record the outcome of an admitted run (salvage kind or null on success). */
   recordOutcome: (fingerprint: string, salvage: BriefSalvageKind | null) => void;
   /**
@@ -153,77 +121,31 @@ export function createBriefDispatchLedger(): BriefDispatchLedger {
 
     admit(fingerprint) {
       const existing = byFingerprint.get(fingerprint);
-      if (existing?.lastSalvage !== undefined && isHardBlockSalvage(existing.lastSalvage)) {
-        return {
-          ok: false,
-          message: hardBlockMessage(existing.lastSalvage, existing.dispatchCount),
-        };
-      }
       const nextCount = (existing?.dispatchCount ?? 0) + 1;
-      byFingerprint.set(fingerprint, {
-        dispatchCount: nextCount,
-        ...(existing?.lastSalvage !== undefined ? { lastSalvage: existing.lastSalvage } : {}),
-      });
+      byFingerprint.set(fingerprint, { dispatchCount: nextCount });
       return { ok: true, dispatchCount: nextCount };
     },
 
     recordOutcome(fingerprint, salvage) {
-      const existing = byFingerprint.get(fingerprint);
-      if (existing === undefined) {
-        // admit() always runs first in production; keep defensive for unit tests.
-        byFingerprint.set(fingerprint, {
-          dispatchCount: salvage === null ? 0 : 1,
-          ...(salvage !== null ? { lastSalvage: salvage } : {}),
-        });
-        return;
-      }
+      // A successful complete resets the same-brief retry budget. Any other
+      // salvage leaves dispatchCount as admit() already recorded it.
       if (salvage === null) {
-        // CL-6710: a successful complete clears the sticky hard-block too.
-        // Two concurrent identical-brief dispatches can both admit; if one
-        // salvages and the other succeeds, the success proves the brief is
-        // re-dispatchable, so it must not leave the sibling's hard-block
-        // standing for the rest of the session.
         byFingerprint.set(fingerprint, { dispatchCount: 0 });
-        return;
       }
-      byFingerprint.set(fingerprint, {
-        dispatchCount: existing.dispatchCount,
-        lastSalvage: salvage,
-      });
     },
 
     release(fingerprint) {
       const existing = byFingerprint.get(fingerprint);
       if (existing === undefined) return;
       if (existing.dispatchCount <= 1) {
-        if (existing.lastSalvage !== undefined) {
-          byFingerprint.set(fingerprint, {
-            dispatchCount: 0,
-            lastSalvage: existing.lastSalvage,
-          });
-        } else {
-          byFingerprint.delete(fingerprint);
-        }
+        byFingerprint.delete(fingerprint);
         return;
       }
-      byFingerprint.set(fingerprint, {
-        dispatchCount: existing.dispatchCount - 1,
-        ...(existing.lastSalvage !== undefined ? { lastSalvage: existing.lastSalvage } : {}),
-      });
+      byFingerprint.set(fingerprint, { dispatchCount: existing.dispatchCount - 1 });
     },
   };
 }
 
-function hardBlockMessage(salvage: HardBlockSalvage, priorDispatches: number): string {
-  return (
-    `Error: refused re-dispatch of an identical task brief after a ${salvage} salvage ` +
-    `(already dispatched ${priorDispatches} time${priorDispatches === 1 ? "" : "s"}). ` +
-    `Change the brief (prompt, agent, intent, success_criteria, and/or do_not) before retrying — ` +
-    `raising maxTurns alone will not unlock this fingerprint. ` +
-    `To force a re-run of the same work, alter at least one of those fields so the fingerprint changes.`
-  );
-}
-
 /**
  * Whether turn-budget parent hint should recommend stopping rather than
  * re-dispatching with a higher maxTurns.