pi-goal-list-loop-audit 0.38.35 → 0.38.37

package/CHANGELOG.md

@@ -1,5 +1,30 @@
 # Changelog
 
+## 0.38.37 — bullets + deliberate non-do (2026-09-09)
+
+### Fixed
+
+- **Auditor disapproval of v0.38.36 answered:** the posted summary now carries one `•` bullet per informing detail on both posted surfaces (chat + transcript); new optional `complete_goal` `leftOut` parameter renders as the closing `• Left out:` bullet on all three approval sites (fresh, detached-retry, Esc-anyway). Absent stays absent — never invented, filler drops. Recap/archive format unchanged.
+- **Stale pins updated deliberately:** approval-chat `deepEqual` expectations now bullet-prefixed.
+
+### Verification
+
+- New pins: bullet count/evidence + present/absent/filler non-do (render unit), end-to-end `leftOut` claim → posted bullet (MockPi).
+- Full `release:check`: 2040 pass / 2 skip / 0 fail across 204 files (`/var/tmp/glla-bullets-release-check2.log`, tarball `pi-goal-list-loop-audit-0.38.37.tgz`); `tsc --noEmit` clean.
+- Evidence: `audit/HUMAN-VOICE-SUMMARY-2026-09-08.md`.
+
+## 0.38.36 — human-voice terminal summaries (2026-09-08)
+
+### Fixed
+
+- **Summaries read as machine receipts (field 20260908_223522/223523):** `clipSummaryValue` now cuts at clause boundaries and strips stranded punctuation (`…append/replace Apply…`, never `…playlist auto-add,…`); the counts line is verdict-only (`— audit: auditor approved (1 verdict).`, raw run stats stay in the archive); the AUDIT PENDING notice states facts instead of agent imperatives. Agent-written summaries get Codex-shape guidance in the continuation prompt (outcome-first, verifiable bullets, clause cuts). Researched from real Codex transcripts + agy binary strings; user chose Codex-style chat shape.
+- **Stale pins updated deliberately:** `— run:` telemetry assertions, `/detached auditor queued/` wordings, continuation-payload byte fixtures (+344/payload, growth exactly linear).
+
+### Verification
+
+- Full `release:check`: 2038 pass / 2 skip / 0 fail across 204 files; `tsc --noEmit` clean.
+- Evidence: `audit/HUMAN-VOICE-SUMMARY-2026-09-08.md`.
+
 ## 0.38.35 — closed card + smart summary (2026-09-08)
 
 ### Fixed

package/docs/INDEX.md

@@ -18,7 +18,7 @@ For shipped docs, the relevant entry points are:
   failback; v0.35.9 hardened cross-version npm tarball checks; v0.35.10
   handles multi-entry npm dry-run reports; v0.35.11 accepts both npm report
   shapes; v0.35.12 supports npm 12's keyed pack reports; v0.35.13 fixes stale-API recovery loops.
-  v0.35.14–v0.38.35 continue through the supervisor freeze (`/glla pause`),
+  v0.35.14–v0.38.37 continue through the supervisor freeze (`/glla pause`),
   load hold, auditor picker parity, Windows launch fix, zombie-watchdog
   subagent carve-out, due-wait backstop, the `/glla agents` visibility panel,
   durable state-root selection, blank-until-resume auditor context, frozen

package/extensions/completion-summary.ts

@@ -78,18 +78,33 @@ export function isUsefulCompletionSummary(text: string | undefined): boolean {
  * Missing labels are retained as `not recorded` so a compact notification
  * cannot accidentally imply evidence that the durable recap did not contain.
  */
-/** Cut a summary value at a word boundary, never mid-word (the
- * Screenshot_20260903_204003/204005 complaint: `0 o…`, `qu…`, `belo…`).
- * Falls back to a hard cut only when the head holds no space past the
- * halfway mark — a long token such as a commit hash must not eviscerate
- * the whole value. Short values pass through untouched (no `…`). */
+/** Cut a summary value at a clause boundary, never mid-word (field
+ * complaints 2026-09-03 `0 o…` and 2026-09-08 `playlist auto-add,…`:
+ * a word-boundary cut that strands dangling punctuation still reads as
+ * clipping, not summarizing). Prefer the last clause boundary
+ * (`, ; : · — – ( [`) past a floor so the cut reads intentional; fall
+ * back to the word break, then to a hard cut only when the head holds
+ * no space past the halfway mark — a long token such as a commit hash
+ * must not eviscerate the whole value. Trailing punctuation is stripped
+ * before the ellipsis. Short values pass through untouched (no `…`). */
 export function clipSummaryValue(value: string, limit: number): string {
   const clean = value.replace(/\s+/g, " ").trim();
   const capped = Number.isFinite(limit) ? Math.max(8, Math.floor(limit)) : 72;
   if (clean.length <= capped) return clean;
   const head = clean.slice(0, capped - 1);
+  const floor = Math.max(16, Math.floor((capped - 1) * 0.4));
+  let boundary = -1;
+  for (let i = 0; i < head.length; i++) {
+    if (/[,;:·—–(\[]/.test(head[i]!)) boundary = i;
+  }
+  if (boundary >= floor) {
+    const cut = head.slice(0, boundary).replace(/[,;:·—–(\[\s]+$/u, "");
+    if (cut.length >= Math.min(floor, 16)) return `${cut}…`;
+  }
   const space = head.lastIndexOf(" ");
-  const kept = (space > capped / 2 ? head.slice(0, space) : head).trimEnd();
+  const kept = (space > capped / 2 ? head.slice(0, space) : head)
+    .trimEnd()
+    .replace(/[,;:·—–(\[]$/u, "");
   return `${kept}…`;
 }
 
@@ -185,22 +200,21 @@ export function buildApprovalChatLines(notice: {
 }): string[] {
   return [
     `✓ done — ${notice.outcome}`,
-    ...withoutStaleNext(notice.details),
+    // v0.38.37 (audit 2026-09-08): one verifiable-result bullet per
+    // informing detail — the Codex closing shape.
+    ...withoutStaleNext(notice.details).map((detail) => `• ${detail}`),
     notice.approval,
     ...(notice.counts ? [notice.counts] : []),
     notice.record,
   ];
 }
 
-/** v0.38.25: the audit-goal counts line. Compact execution + audit proof
- * built ONLY from durable goal state — turns/file-writes/bash from
- * telemetry, verdict counts from auditHistory. Absent facts are named as
- * absent, never invented (the recorded-facts honesty rule). */
+/** v0.38.25: the audit-goal counts line — verdict proof ONLY, built from
+ * durable goal state. Raw run stats (turns/file-writes/bash calls) read as
+ * a machine receipt in user chat (field 2026-09-08 223522/223523); the
+ * full execution evidence stays in the six-label archive record. Absent
+ * facts are named as absent, never invented (recorded-facts honesty). */
 export function buildAuditCountsLine(goal: Goal, auditNote?: string): string {
-  const telemetry = goal.telemetry;
-  const run = telemetry
-    ? `${telemetry.turns} turns · ${telemetry.fileWrites} file writes · ${telemetry.bashCalls} bash calls`
-    : "no execution telemetry was recorded";
   const history = goal.auditHistory ?? [];
   const latest = history.length > 0 ? history[history.length - 1] : undefined;
   const audit = auditNote ?? (latest === undefined
@@ -209,7 +223,7 @@ export function buildAuditCountsLine(goal: Goal, auditNote?: string): string {
       const verdict = latest.approved ? "approved" : latest.impossible ? "impossible" : latest.disapproved ? "disapproved" : "no verdict";
       return `auditor ${verdict} (${history.length} verdict${history.length === 1 ? "" : "s"})`;
     })());
-  return `— run: ${run} · ${audit}.`;
+  return `— audit: ${audit}.`;
 }
 
 export interface TerminalApprovalRenderInput {
@@ -228,6 +242,8 @@ export interface TerminalApprovalRenderInput {
   countsLine?: string;
   /** Override for the audit half of the counts line. */
   auditNote?: string;
+  /** v0.38.37: what the agent deliberately left out (complete_goal leftOut). */
+  leftOut?: string;
 }
 
 export interface TerminalApprovalRender {
@@ -259,7 +275,15 @@ export function buildTerminalApprovalRender(input: TerminalApprovalRenderInput):
   };
   const candidate = input.completionSummary ?? input.goal.completionSummary;
   const recap = compactTerminalCompletionSummary(facts, candidate);
-  const brief = terminalHumanBrief(facts, candidate);
+  const baseBrief = terminalHumanBrief(facts, candidate);
+  // v0.38.37 (audit 2026-09-08): the deliberate non-do comes from the
+  // agent's complete_goal leftOut claim — never invented. Filler ("none")
+  // drops via the same briefValueContent filter; absent stays absent.
+  const leftOut = input.leftOut?.trim();
+  const leftOutContent = leftOut ? briefValueContent(leftOut) : null;
+  const brief = leftOutContent
+    ? { ...baseBrief, details: [...baseBrief.details, `Left out: ${clipSummaryValue(leftOutContent, 120)}`] }
+    : baseBrief;
   const countsLine = input.countsLine ?? buildAuditCountsLine(input.goal, input.auditNote);
   return {
     chatLines: [
@@ -273,7 +297,7 @@ export function buildTerminalApprovalRender(input: TerminalApprovalRenderInput):
       ...(input.extras ?? []),
     ],
     recap,
-    transcriptLines: [...withoutStaleNext(brief.details), input.approval],
+    transcriptLines: [...withoutStaleNext(brief.details).map((detail) => `• ${detail}`), input.approval],
     countsLine,
     outcome: brief.outcome,
     approval: input.approval,

package/extensions/goal-loop-core.ts

@@ -280,6 +280,12 @@ export interface PendingCompletion {
   recoveryEpisodeKey?: string;
   /** Durable per-episode notice fence; display projections must consult it. */
   recoveryNoticeKeys?: string[];
+  /**
+   * v0.38.37: what the agent deliberately left out of this turn's work
+   * (complete_goal leftOut). Renders as the closing non-do bullet in the
+   * terminal user summary; absent means nothing was deliberately left out.
+   */
+  leftOut?: string;
   /**
    * Durable one-shot recovery fence. A parked claim may receive one
    * automatic retry after a validated healthy lifecycle/recovery event;

package/extensions/loops/goal-auditor-hooks.ts

@@ -1439,6 +1439,8 @@ async function retryStoredCompletionAudit(origin: CompletionAuditOrigin = "provi
       completionSummary: state.goal.completionSummary,
       approval: `— auditor ${result.model} approved${approvalVia}.`,
       record: approvalRecord,
+      // v0.38.37: the deliberate non-do the agent claimed, if any.
+      ...(claim.leftOut ? { leftOut: claim.leftOut } : {}),
       extras: inspectionSessionPath
         ? [`Auditor session kept for review: pi --session ${inspectionSessionPath} (or pi --fork ${inspectionSessionPath}).`]
         : [],

package/extensions/loops/goal-tools.ts

@@ -519,6 +519,13 @@ function registerAgentTools(pi: any): void {
           "and cross-checks the verification claim against real artifacts (see audit/COMPLETION-SUMMARY-POLICY-2026-08-19.md).",
       })),
       newObjective: Type.Optional(Type.String({ description: "v0.25.0 (contract item 15): when the work has legitimately shifted, pass the new objective here — it atomically replaces the goal objective AND the audit proceeds against the NEW objective in this same call. Do not use to dodge a legitimate disapproval; the auditor sees the change." })),
+      leftOut: Type.Optional(Type.String({
+        maxLength: 500,
+        description:
+          "v0.38.37: what this turn deliberately left out (scope cut, deferred item + why, in plain words). " +
+          "Renders as the closing `• Left out:` bullet in the user-facing terminal summary. " +
+          "Omit when nothing was deliberately left out — absent stays absent, never invented.",
+      })),
     }),
     async execute(_id, params, signal, _onUpdate, execCtx) {
       const foreign0 = foreignToolGuard(execCtx);
@@ -542,7 +549,7 @@ function registerAgentTools(pi: any): void {
         }
         return { content: [{ type: "text", text: `No active goal — it is ${state.goal.status}.` }], details: {} };
       }
-      const p = params as { completionSummary?: string; verificationSummary?: string; newObjective?: string };
+      const p = params as { completionSummary?: string; verificationSummary?: string; newObjective?: string; leftOut?: string };
       if (state.goal.repairTarget) {
         return {
           content: [{ type: "text", text: `This repair card cannot be completed yet. Redraft the original target as a confirmed task list with propose_task_list (include objective: ${state.goal.repairTarget.objective.slice(0, 180)}), then continue the real work.` }],
@@ -742,6 +749,9 @@ function registerAgentTools(pi: any): void {
       const completionClaim = beginCompletionAudit(ctx, {
         completionSummary: finalSummary,
         verificationSummary: p.verificationSummary,
+        // v0.38.37: the deliberate non-do rides the pending claim into
+        // the terminal render — the only source the summary may cite.
+        ...(p.leftOut?.trim() ? { leftOut: p.leftOut.trim().slice(0, 500) } : {}),
         at: nowIso(),
       }, "complete-goal");
       if (!completionClaim) {
@@ -1194,6 +1204,9 @@ function registerAgentTools(pi: any): void {
       // Escape hatch: the user aborted the audit (Esc). Offer the explicit
       // choice — complete WITHOUT audit, or keep working. (pi-goal-x parity.)
       if (result.error === "Auditor aborted.") {
+        // v0.38.37: capture the agent-claimed non-do before the claim is
+        // cleared below — the Esc-complete path renders the same voice.
+        const escLeftOut = state.goal.pendingCompletion?.leftOut;
         updateGoal({ status: "active", auditHistory: history, pendingCompletion: undefined, pauseReason: "audit aborted by user (Esc)" }, ctx);
         const abortConfirmCtx = freshCtxForGeneration(auditGeneration);
         if (!abortConfirmCtx) return staleToolResult();
@@ -1232,6 +1245,9 @@ function registerAgentTools(pi: any): void {
             approval: `— completed without audit (your choice).`,
             record: `— record: ${escArchivePath}`,
             auditNote: "completed without audit (your choice)",
+            // v0.38.37: the claim was cleared pre-confirm; the non-do was
+            // captured from it above.
+            ...(escLeftOut ? { leftOut: escLeftOut } : {}),
           });
           if (!archiveCurrentGoal(ctx, "complete", terminalReason)) {
             return {
@@ -1286,6 +1302,8 @@ function registerAgentTools(pi: any): void {
           completionSummary: state.goal.completionSummary,
           approval: `— auditor ${result.model} approved.`,
           record: manualArchiveRecord,
+          // v0.38.37: the deliberate non-do rides the durable claim.
+          ...(durableCompletionClaim.leftOut ? { leftOut: durableCompletionClaim.leftOut } : {}),
           extras: inspectionSessionPath
             ? [`Auditor session kept for review: pi --session ${inspectionSessionPath} (or pi --fork ${inspectionSessionPath}).`]
             : [],
@@ -1876,7 +1894,7 @@ function registerAgentTools(pi: any): void {
         if (notifyFailure) current.ui.notify(`Completion auditor worker failed to settle (infrastructure, not a verdict). The stored claim is safe; ${activeGoalSurfaceCommand("resume")} retries it.`, "warning");
       });
       return {
-        content: [{ type: "text", text: `AUDIT PENDING — nonterminal. Completion claim persisted; detached auditor queued (model: ${via ?? "setting"}). The goal is not approved or complete. Do not claim completion or give a final success summary. Do not wait or poll: GLLA will post the concrete final summary after verified approval and durable archive; rejection resumes work.` }],
+        content: [{ type: "text", text: `AUDIT PENDING — nonterminal. The completion claim is stored and a detached auditor is settling it (model: ${via ?? "setting"}). The goal is not approved or complete yet. The final summary posts here after verified approval and durable archive; rejection resumes work automatically. Ending this turn now is correct — no waiting or polling needed.` }],
         details: { status: "audit-pending", terminal: false },
         // End this tool batch, not the goal; the detached worker owns settlement.
         terminate: true,

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "pi-goal-list-loop-audit",
-  "version": "0.38.35",
+  "version": "0.38.37",
   "description": "Mission control for autonomous pi: interview-drafted goals, an audited task queue, and forever-loops (metric, spec, project-audit) that run for hours. A detached extension-less auditor process re-verifies every completion with raw evidence without holding the main pi turn; confirmed drafts, decision pauses and consent gates keep you in charge.",
   "license": "AGPL-3.0-only",
   "author": "dracon",

package/prompts/goal-loop-continuation.md

@@ -196,3 +196,5 @@ The orchestrator's backstop is the stall watchdog: three consecutive turns with
 ### Completion communication
 
 `complete_goal` submits a nonterminal claim, not a completion verdict. While the detached audit is pending, do not say the goal is done/complete/approved or present a final success summary. Do not wait or poll. If a response is needed, give one brief pending-status sentence. GLLA posts the concrete outcome-first summary into chat after verified approval and durable archive, without needing another user prompt. Do not duplicate that summary. A rejected claim remains unfinished and repair work continues.
+
+The agent-written completionSummary is user-facing evidence, not a step log: open with the outcome in plain words, then one verifiable result per label with its proof inline (test counts, file paths, commit refs) — never raw run stats. Cut values at clause boundaries, never mid-word; pass what was deliberately left out as the complete_goal leftOut parameter — it renders as the closing bullet, and is omitted when absent.