@ai-dev-methodologies/rlp-desk 0.16.0 → 0.18.0

package/CHANGELOG.md

@@ -11,6 +11,60 @@ For pre-v0.15.4 versions, refer to `git log` and individual GitHub release notes
 - Post-v0.15.6: remove `RLP_LIFECYCLE_METRICS` flag entirely (per plan v3 ADR follow-ups).
 - Phase D.1 (handoff documents) + Phase D.2 (per-stage agent role specialization) — both deferred per `docs/plans/v0.15.4-release-runbook.md` §7.6.
 
+## [0.18.0] — 2026-06-24
+
+**Leader hardening: campaigns that complete.** The `--mode tmux` leader's decision
+gates used to TERMINATE where they should RECOVER, so campaigns could stall at
+`--max-iter` or block on a single transient slip. This release resolves that
+"never completes" failure class.
+
+### Fixed
+- A single transient `blocked` verdict no longer ends the campaign. A fresh-context
+  Worker/Verifier mis-emitting `blocked` once now gets grace (soft-fail + retry);
+  termination only on a genuine infra failure or a repeated/consecutive block.
+- Verifier phrasing variants (`completed`/`done`, lowercase `all`) no longer strand a
+  genuinely-complete campaign at the iteration cap — recommendation/us_id normalized at read.
+- Mixed-engine dead-pane detection: a dead pane is detected per role when Worker and
+  Verifier run different engines (e.g. claude worker + codex verifier).
+- No double-credit of a re-submitted User Story on the verify pass-path.
+- Final verification now retries a single transient poll failure instead of charging a
+  false fail at the end of the campaign.
+- Durable sentinels: atomic writes detect truncation (disk-full/SIGPIPE) instead of
+  renaming a half-written file into the canonical path.
+- Race-safe locks: the runner-lock and per-slug lock close an ABA/empty-owner race that
+  could let two leaders run on one project root.
+- Done-claim freshness gate: auto-synthesis rejects a stale/wrong-US done-claim, preventing
+  the wrong User Story from being credited.
+- Model-upgrade state survives relaunch: a resumed campaign keeps the upgraded worker model
+  instead of resetting to the CLI default.
+- codex 0.141 worker compatibility: the directory-trust prompt is accepted and MCP disabled
+  so codex workers don't block on startup.
+
+### Changed
+- Stronger final verification: `--final-verifier-model` / `--final-verifier-effort` are now
+  actually used for the final ALL-verify pass (previously every verify ran at the per-US
+  model). "Final = stricter" now holds.
+- Consensus cross-verifier model knobs now take effect: `--consensus-model` (per-US, lighter)
+  and `--final-consensus-model` (final, stricter) are wired into the consensus path —
+  previously documented but inert.
+- Verifier FORMAT meta-gates softened cross-engine while correctness gates stay strict
+  (fewer false fails on phrasing, no weakening of substance).
+
+## [0.17.0] — 2026-06-19
+
+**BREAKING (ADR-001): `node run.mjs run <slug> --mode agent` (the deprecated Node-leader direct-CLI alpha) now hard-errors.** Per the schedule announced in 0.16.0, the `--mode agent` entry point exits 2 with a redirect, and the Node-CLI default mode flips from `agent` to `tmux` (the canonical production leader). The `src/node/**` engine modules are retained (they back the Native Agent() path and the test suite); only the direct-CLI `--mode agent` *dispatch entry* was removed. **Migration:** replace `node run.mjs run <slug> --mode agent` with `--mode tmux`. The slash command's `--mode native` default is unaffected.
+
+### Changed
+- `node run.mjs run <slug>` (no `--mode`) now defaults to `--mode tmux` and delegates to the zsh leader, instead of the deprecated Node leader.
+
+### Fixed
+- Four shipped files referenced `docs/multi-mission-orchestration.md`; corrected to the real shipped path `docs/rlp-desk/multi-mission-orchestration.md` (dead link on fresh installs). Added a docs-link-resolve test.
+- `uninstall` now removes the `UNLOCK.md` marker (and unlocks 0o444 files before unlink), so the `ralph-desk/` directory is fully removed.
+- The A4 fallback iteration-signal write now goes through the atomic `atomic_write` helper (temp+rename) instead of a raw redirect.
+
+### Changed (packaging)
+- The shipped `examples/calculator` scaffold moved from the legacy `.claude/ralph-desk/` path to `.rlp-desk/` (v0.13.0+ convention) and no longer ships generated runtime state.
+
 ## [0.16.0] — 2026-06-18
 
 Leader consolidation (ADR-001): `--mode tmux` is now the canonical production leader, self-verification works on it, and the deprecated `--mode agent` Node-CLI path is on a dated removal schedule.

package/docs/rlp-desk/getting-started.md

@@ -83,7 +83,7 @@ This creates the scaffold:
 
 ## Step 5: Customize the PRD
 
-Edit `.rlp-desk/plans/prd-loop-test.md` to define your user stories and acceptance criteria. See [`examples/calculator/`](../examples/calculator/.claude/ralph-desk/plans/prd-loop-test.md) for a complete example.
+Edit `.rlp-desk/plans/prd-loop-test.md` to define your user stories and acceptance criteria. See [`examples/calculator/`](../examples/calculator/.rlp-desk/plans/prd-loop-test.md) for a complete example.
 
 Key sections:
 - **User Stories** with Given/When/Then acceptance criteria, Task Type, and Risk Level

package/docs/rlp-desk/protocol-reference.md

@@ -649,5 +649,5 @@ an optional `next_mission_candidate` field:
 
 The leader propagates the field into `status.json`
 (`status.next_mission_candidate`). Wrappers poll either source. See
-`docs/multi-mission-orchestration.md` for the consumer-side polling
+`docs/rlp-desk/multi-mission-orchestration.md` for the consumer-side polling
 pattern.

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@ai-dev-methodologies/rlp-desk",
-  "version": "0.16.0",
+  "version": "0.18.0",
   "description": "Fresh-context iterative loops for Claude Code — autonomous task completion with independent verification",
   "scripts": {
     "postinstall": "node scripts/postinstall.js",
@@ -23,10 +23,10 @@
     "scripts/",
     "docs/rlp-desk/*.md",
     "docs/rlp-desk/blueprints/",
-    "examples/calculator/.claude/ralph-desk/context/",
-    "examples/calculator/.claude/ralph-desk/memos/",
-    "examples/calculator/.claude/ralph-desk/plans/",
-    "examples/calculator/.claude/ralph-desk/prompts/",
+    "examples/calculator/.rlp-desk/context/",
+    "examples/calculator/.rlp-desk/memos/",
+    "examples/calculator/.rlp-desk/plans/",
+    "examples/calculator/.rlp-desk/prompts/",
     "install.sh",
     "README.md",
     "CHANGELOG.md",

package/scripts/uninstall.js

@@ -25,10 +25,22 @@ const files = [
   path.join(deskDir, "init_ralph_desk.zsh"),
   path.join(deskDir, "run_ralph_desk.zsh"),
   path.join(deskDir, "lib_ralph_desk.zsh"),
+  // PKG-6: postinstall writes the UNLOCK.md escape-hatch doc into deskDir. If we
+  // skip it here, the empty-dir rmdir below always sees one leftover file and the
+  // ralph-desk/ directory is never removed.
+  path.join(deskDir, "UNLOCK.md"),
 ];
 
 for (const targetPath of files) {
   try {
+    // Installed files are write-locked (chmod 0o444) per postinstall v5.7 §4.10.
+    // Restore write permission before unlink so removal is robust across
+    // filesystems (mirrors postinstall's unlock-before-remove).
+    try {
+      fs.chmodSync(targetPath, 0o644);
+    } catch (_) {
+      // File may be missing or already writable.
+    }
     fs.rmSync(targetPath, { recursive: true, force: true });
     console.log("  - " + targetPath);
   } catch (_) {

package/src/commands/rlp-desk.md

@@ -374,7 +374,7 @@ This contract MUST be observed in every iteration of the leader loop below. Futu
 
 #### Direct Node CLI invocation (`node run.mjs run <slug> --mode agent` — deprecated alpha)
 
-Direct invocation of `node ~/.claude/ralph-desk/node/run.mjs run <slug> --mode agent` is **the deprecated Node-leader alpha path**. This is unrelated to the slash command's Native Agent() path above — different code, different leader, different lifecycle. The Node leader currently retains SV/flywheel implementations not yet ported to Native Agent(). For production tmux orchestration, use `--mode tmux` (the **canonical** leader). For Claude Code Native Agent() campaigns, use `/rlp-desk run <slug> --mode native` from a Claude Code session.
+Direct invocation of `node ~/.claude/ralph-desk/node/run.mjs run <slug> --mode agent` **hard-errors (exit 2) as of this release** (per [ADR-001](../../docs/plans/adr-001-leader-consolidation.md) §3) — it was the deprecated Node-leader alpha path, and the dated breaking change has now landed. Direct CLI invocation redirects to `--mode tmux` (production) / `--mode native` (slash), and the **Node-CLI default mode is now `tmux`**. This is unrelated to the slash command's legacy `--mode agent` → native redirect — different code, different leader, different lifecycle. The `src/node/**` engine modules (which still back the Native Agent() path and the test suite, and retain SV/flywheel implementations not yet ported to Native Agent()) are **retained** — only the direct-CLI `--mode agent` *entry point* hard-errors. For production tmux orchestration, use `--mode tmux` (the **canonical** leader). For Claude Code Native Agent() campaigns, use `/rlp-desk run <slug> --mode native` from a Claude Code session.
 
 **Deprecation schedule** (per [ADR-001](../../docs/plans/adr-001-leader-consolidation.md) §3 — applies to the Node-CLI `--mode agent` entry point ONLY; the `src/node/**` engine modules are retained throughout):
 

package/src/governance.md

@@ -66,9 +66,25 @@ Verification layers:
 - L3: E2E Simulation — known input → full pipeline → output comparison. Always required.
 - L4: Deploy Verify — production environment checks. Required when deploying.
 
-Layer requirements per US are determined by risk classification (§1c).
-Non-applicable layers must be explicitly marked "N/A — {reason}" in the test-spec.
-Any required layer section with TODO or blank = automatic Verifier FAIL.
+Layer requirements per US are determined by risk classification (§1c). Layer
+completeness means ACTUAL verification COVERAGE, not test-spec FORMAT: a required
+layer is satisfied when its behavior is genuinely exercised by a concrete, PASSING
+check — e.g. a per-AC command in the Criteria→Verification table counts as L1/L3
+coverage. Explicit "## L1/L2/L3" section headers and "N/A — {reason}" markers are
+recommended but NOT mandatory, and their absence is NOT itself a fail.
+FAIL a required layer ONLY when its verification is genuinely absent, blank, TODO,
+or failing — never merely for a different-but-adequate test-spec format, a missing
+N/A marker, or RED evidence aggregated across the US instead of recorded per-AC.
+The iter-signal.json only identifies WHICH US to verify; whether the Worker wrote
+it or the leader synthesized it does not change the verdict — verify the work, not
+the signal's author. Deliverable COMPLETENESS is NOT exempt: if the work an AC
+requires is absent, uncommitted/untracked, or never actually exercised, the layer
+still FAILs. This rule is IDENTICAL for every verifier engine (claude AND codex):
+do NOT block a PASS on test-spec FORMAT when the acceptance criteria are met and
+their checks are green — but DO fail on missing or unverified substance. This
+preserves strict AC + real layer coverage (IL-4 still governs test sufficiency);
+it removes only format pedantry, never substance — the F-17 cross-engine
+consistency fix (scope narrowed to format-only per the F-18 review).
 See §1d for full layer definitions.
 
 **IL-4: Test Sufficiency**
@@ -584,7 +600,7 @@ for iteration in 1..max_iter:
   ⑥½ Flywheel direction review (when --flywheel on-fail and consecutive_failures > 0)
      - Dispatch Flywheel agent (fresh context, --flywheel-model)
      - Read flywheel-signal.json for direction decision (hold/pivot/reduce/expand)
-     - Optional `next_mission_candidate` field (string | null): when present, the leader propagates it to status.json so consumer wrappers can chain the next mission without code edits. See docs/multi-mission-orchestration.md.
+     - Optional `next_mission_candidate` field (string | null): when present, the leader propagates it to status.json so consumer wrappers can chain the next mission without code edits. See docs/rlp-desk/multi-mission-orchestration.md.
      - If --flywheel-guard on:
        - Dispatch Guard agent (fresh context, --flywheel-guard-model)
        - Read flywheel-guard-verdict.json:

package/src/node/run.mjs

@@ -14,7 +14,11 @@ import {
 import { isClaudeEngine } from './cli/command-builder.mjs';
 
 const RUN_DEFAULTS = {
-  mode: 'agent',
+  // ARCH Wave D (ADR-001 §3): the Node-CLI default mode flips from the deprecated
+  // 'agent' (Node-leader alpha) to 'tmux' (the canonical production leader) at the
+  // same step that makes --mode agent hard-error. A bare `run <slug>` now delegates
+  // to the zsh runner, not the deprecated Node leader.
+  mode: 'tmux',
   workerModel: 'haiku',
   verifierModel: 'sonnet',
   finalVerifierModel: 'opus',
@@ -49,14 +53,14 @@ function buildHelpText() {
     'Commands:',
     '  brainstorm <description>     Plan before init (not implemented in the Node rewrite yet)',
     '  init <slug> [objective]      Create project scaffold',
-    '  run <slug> [options]         Run loop (tmux=zsh leader [production], agent=Node leader [deprecated alpha], native=slash-only error)',
+    '  run <slug> [options]         Run loop (tmux=zsh leader [production, default], agent=hard-errors per ADR-001, native=slash-only error)',
     '  status <slug>                Show loop status',
     '  logs <slug> [N]              Show iteration log (not implemented in the Node rewrite yet)',
     '  clean <slug> [--kill-session] Reset for re-run (removes sentinels + runtime/; preserves PRD/prompts/memory)',
     '  resume <slug>                Resume loop (not implemented in the Node rewrite yet)',
     '',
     'Run Options:',
-    '  --mode tmux|agent|native       (CLI: tmux=production, agent=deprecated, native=errors with redirect to slash command)',
+    '  --mode tmux|agent|native       (CLI: tmux=production [default], agent=hard-errors (ADR-001), native=errors with redirect to slash command)',
     '  --worker-model MODEL',
     '  --lock-worker-model',
     '  --verifier-model MODEL',
@@ -373,7 +377,7 @@ async function runTmuxViaZsh(slug, options, deps) {
   if (unsupported.length > 0) {
     write(
       deps.stderr,
-      `WARNING: ${unsupported.join(', ')} not honored in --mode tmux (zsh runner). Flywheel is deprecated (ADR-001) — use --mode agent if you still need it.`,
+      `WARNING: ${unsupported.join(', ')} not honored in --mode tmux (zsh runner). Flywheel is deprecated (ADR-001) and unimplemented in the canonical leader.`,
     );
   }
 
@@ -442,7 +446,7 @@ async function runRunCommand(args, deps) {
     );
     write(
       deps.stderr,
-      'If hang persists, switch to --worker-model gpt-5.5:high (codex) or --mode agent.',
+      'If hang persists, switch to --worker-model gpt-5.5:high (codex).',
     );
   }
 
@@ -468,50 +472,38 @@ async function runRunCommand(args, deps) {
     );
     write(
       deps.stderr,
-      'or use `--mode tmux` (production) / `--mode agent` (deprecated alpha) for direct CLI invocation.',
+      'or use `--mode tmux` (production) for direct CLI invocation. `--mode agent` hard-errors (ADR-001).',
     );
     return 2;
   }
 
-  // P1.b: --mode agent (Node-leader alpha) is deprecated. The slash command's
-  // Native Agent() path (`/rlp-desk run --mode native`) is unrelated — different
-  // code, different leader. We keep the Node-leader behavior unchanged for
-  // backward compatibility but surface a strong deprecation banner so wrappers
-  // can migrate before the next major release hard-errors this mode.
-  if (
-    options.mode === 'agent'
-    && !process.env.RLP_DESK_QUIET_WARNINGS
-    && process.env.NODE_ENV !== 'test'
-  ) {
-    write(
-      deps.stderr,
-      'WARNING: --mode agent (Node-leader alpha) is deprecated.',
-    );
+  // ARCH Wave D (ADR-001 §3): --mode agent (Node-leader direct-CLI alpha) HARD-ERRORS
+  // as of this release. It is the dated breaking change the deprecation banner
+  // announced — direct CLI invocation now exits 2 with a redirect to the canonical
+  // production leader (--mode tmux) or the slash-command Native Agent() path
+  // (--mode native). This is UNRELATED to the slash command's legacy `--mode agent`
+  // → native redirect (different code, different leader). The `src/node/**` engine
+  // modules (run()/runCampaign) are RETAINED — they remain the engine the Native
+  // Agent() path and the test suite build on; only this direct-CLI dispatch entry
+  // hard-errors. Mirror the --mode native slash-only exit-2 pattern above.
+  if (options.mode === 'agent') {
     write(
       deps.stderr,
-      'This is the direct Node-CLI alpha path — UNRELATED to the slash command Native Agent() path (`/rlp-desk run --mode native`).',
+      'ERROR: --mode agent (Node-leader direct-CLI alpha) is no longer supported (ADR-001).',
     );
     write(
       deps.stderr,
-      'For production tmux orchestration, use `--mode tmux`.',
+      'For production tmux orchestration, use `--mode tmux` (the canonical leader).',
     );
     write(
       deps.stderr,
-      'For Claude Code Native Agent() campaigns, use `/rlp-desk run --mode native` from a Claude Code session.',
+      'For Claude Code Native Agent() campaigns, use `/rlp-desk run <slug> --mode native` from a Claude Code session.',
     );
-    // 2026-05-07 (v0.15.2): rlp-desk is in active stabilization. Goal: reach
-    // omc /team/ralph/ralplan level of reliability while preserving
-    // rlp-desk's self-driving advantages (multi-engine consensus, multi-mission
-    // queue, BLOCK_TAGS taxonomy, structured SV reports). omc is the BENCHMARK,
-    // not a replacement. See docs/plans/v0.15-stabilization-plan.md.
     write(
       deps.stderr,
-      'SCHEDULED REMOVAL (ADR-001): --mode agent (Node CLI alpha) hard-errors in 0.17.0 (Node-CLI default flips to tmux) and the dispatch branch is removed in 0.18.0. Engine modules stay. Migrate to --mode tmux.',
-    );
-    write(
-      deps.stderr,
-      'STABILIZATION IN PROGRESS: rlp-desk is hardening against the 10-bug regression pattern observed 2026-05-01..05-07. See docs/plans/v0.15-stabilization-plan.md.',
+      'The src/node/** engine modules are retained; only the direct-CLI --mode agent entry point was removed.',
     );
+    return 2;
   }
 
   const result = await deps.runCampaign(slug, options);

package/src/node/runner/campaign-main-loop.mjs

@@ -1715,7 +1715,7 @@ async function _runCampaignBody(slug, options, paths, rootDir) {
       // P0-A multi-mission orchestration: optionally captured from flywheel signal.
       // null when the flywheel did not suggest a next mission. Consumer wrappers
       // poll status.next_mission_candidate to chain missions without code edits.
-      // See docs/multi-mission-orchestration.md.
+      // See docs/rlp-desk/multi-mission-orchestration.md.
       state.next_mission_candidate = flywheelSignal.next_mission_candidate ?? null;
       // Bug #7 Fix-R cleanup: unlock before unlink so 0o444 doesn't block.
       await unlockSentinelFile(paths.flywheelSignalFile);