ralphctl 0.12.1 → 0.13.1

package/dist/manifest.json

@@ -1,6 +1,6 @@
 {
   "version": 1,
-  "generatedAt": "2026-06-14T19:07:40.295Z",
+  "generatedAt": "2026-06-23T21:08:54.874Z",
   "assets": [
     "prompts/_partials/conventions-agents-md.md",
     "prompts/_partials/conventions-claude-md.md",

package/dist/prompts/evaluate/template.md

@@ -136,6 +136,39 @@ from the task description, (3) plan which shell commands to run first.
 
 ## Review protocol
 
+### Phase 0 — Checkpoint write (do this first, before any verification)
+
+Write `signals.json` now with placeholder verdicts — `status: "failed"`, all four floor dimensions
+present, each set to `passed: false` with `finding: "assessment in progress"`. Use the schema and
+path shown in the output contract section at the bottom of this prompt.
+
+This preliminary write is NOT your final verdict. You will overwrite the file with the real verdict
+after Phase 4. Writing it first ensures the harness can recover via corrective retry if this session
+exhausts its token budget mid-analysis — a session that runs out during Phases 1–3 leaves a valid
+`signals.json` on disk rather than a missing one, allowing the harness to prompt a cheaper follow-up
+rather than restarting from scratch.
+
+```json
+{
+  "schemaVersion": 1,
+  "signals": [
+    {
+      "type": "evaluation",
+      "status": "failed",
+      "dimensions": [
+        { "dimension": "correctness", "passed": false, "finding": "assessment in progress" },
+        { "dimension": "completeness", "passed": false, "finding": "assessment in progress" },
+        { "dimension": "safety", "passed": false, "finding": "assessment in progress" },
+        { "dimension": "consistency", "passed": false, "finding": "assessment in progress" }
+      ],
+      "timestamp": "<ISO-8601 timestamp>"
+    }
+  ]
+}
+```
+
+Write this file, then proceed to Phase 1.
+
 ### Phase 1 — Computational verification
 
 Open with a thinking block: list the criteria you will grade and any red flags from the task description.

package/dist/prompts/evaluate-continuation/template.md

@@ -64,6 +64,16 @@ For the complete history — older than the excerpt above — read `{{PROGRESS_F
 {{GENERATOR_HINTS_SECTION}}
 
 <protocol>
+**Checkpoint write — do this first, before re-grading**
+
+If you have not already written a checkpoint `signals.json` for this round, write one now before
+continuing. Use `status: "failed"`, all four floor dimensions present, each set to
+`passed: false` with `finding: "assessment in progress"`. Use the path named in the output
+contract section at the bottom of this prompt. This placeholder is valid against the schema the
+harness validates — it ensures a recoverable file exists on disk if this session exhausts its
+token budget before you reach your final verdict. You will overwrite it after completing all steps
+below.
+
 Re-grade this round the same way you graded the first:
 
 1. Re-run each `auto` criterion's command directly and record the verbatim output. Do NOT run the

package/dist/skills/ralphctl-minimal-scaffolding/SKILL.md

@@ -10,7 +10,7 @@ description: Cross-phase skill — question every harness component on every mod
 > Remove one component at a time when simplifying. Re-examine entire harness when new model releases; strip
 > non-load-bearing pieces."
 >
-> — Anthropic, _Harness Design for Long-Running Apps_
+> — Anthropic, [_Harness Design for Long-Running Application Development_](https://www.anthropic.com/engineering/harness-design-long-running-apps)
 
 Harness complexity drifts upward. Each component that solves a real problem at the time of its addition
 becomes a permanent fixture — even after the model capability that made it necessary has improved past the

package/dist/skills/ralphctl-surgical-simplicity/SKILL.md

@@ -5,8 +5,10 @@ description: Execute-phase skill — write the minimum code the task needs and t
 
 # Surgical Simplicity
 
-> Distilled from Andrej Karpathy's public guidance on LLM coding — his January 2026 X post on coding
-> pitfalls and the "Software Is Changing" / Software 3.0 talk. Clean-room — concepts only, not copied text.
+> Distilled from Andrej Karpathy's public guidance on LLM coding — his January 2026 [X post on coding
+> pitfalls](https://x.com/karpathy/status/2015883857489522876) and the ["Software Is Changing" / Software 3.0
+> talk](https://www.ycombinator.com/library/MW-andrej-karpathy-software-is-changing-again). Clean-room —
+> concepts only, not copied text.
 
 The two failure modes that make AI-generated diffs hard to review are opposite in feel but identical in
 cost: writing too much (speculative code that nobody asked for) and touching too much (sweeping the

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "ralphctl",
-  "version": "0.12.1",
+  "version": "0.13.1",
   "description": "Agent harness for long-running AI coding tasks — orchestrates Claude Code, GitHub Copilot, and OpenAI Codex across repositories",
   "homepage": "https://github.com/lukas-grigis/ralphctl",
   "type": "module",
@@ -49,7 +49,7 @@
   "devDependencies": {
     "@eslint/js": "^10.0.1",
     "@types/cross-spawn": "^6.0.6",
-    "@types/node": "^25.8.0",
+    "@types/node": "^26.0.0",
     "@types/proper-lockfile": "^4.1.4",
     "@types/react": "^19.2.14",
     "@vitest/coverage-v8": "^4.1.6",
@@ -79,6 +79,7 @@
   "scripts": {
     "build": "tsup && tsx scripts/build-assets.ts",
     "dev": "NODE_OPTIONS=--max-old-space-size=8192 tsx src/index.ts",
+    "mock": "tsx scripts/seed-mock.ts",
     "dev:heap-snapshot": "mkdir -p .diagnostics && NODE_OPTIONS='--max-old-space-size=8192 --heapsnapshot-near-heap-limit=2 --diagnostic-dir=.diagnostics' tsx src/index.ts",
     "start": "NODE_OPTIONS=--max-old-space-size=8192 tsx src/index.ts",
     "typecheck": "tsc",