@llblab/pi-kit 0.1.0

package/node_modules/@llblab/pi-actors/skills/project-work/SKILL.md

@@ -0,0 +1,37 @@
+---
+name: project-work
+description: Use for repository health inspection, project summaries, documentation maintenance, release-readiness evidence, or bounded run-operation reports.
+---
+
+# Project Work
+
+Use this Skill to produce bounded project evidence and plans. For generic Recipe execution, Run lifecycle, persistence, or multi-actor methodology, follow `actors` and, when applicable, `swarm`; this Skill only selects project workflows.
+
+## Choose the primary workflow
+
+| Intended result | Recipe | Boundary |
+| --- | --- | --- |
+| Repository status, recent history, docs surface, validation, and risks | `project-work/repo-health` | Runs the caller-supplied trusted validation command |
+| Documentation consistency review and maintenance plan | `project-work/docs-maintenance` | Evidence and plan only; does not edit documentation |
+| Multi-lens release verdict with blockers and degraded-confidence evidence | `project-work/release-readiness` | Readiness only; does not publish |
+| Evidence-only release summary and PR-body draft | `project-work/release-summary` | No commit, PR, merge, tag, publish, or external release action |
+| Bounded report over existing actor Run state | `project-work/run-ops` | Read-only report; does not send Control or mutate Runs |
+
+Choose `release-readiness` when independent review and a release verdict are needed. Choose `release-summary` when the evidence is already sufficient and only a concise operator-gated summary is wanted. If one primary workflow should become a recurring named tool, use the persistent-capability workflow in `actors`; do not copy its composition.
+
+## Inputs and evidence
+
+- Scope every workflow to the caller-selected repository, docs directory, or Run state.
+- Treat `validation_command` as trusted executable input. Do not invent or broaden it.
+- Select current model policy explicitly for workflows that request a model; release-readiness reviewer roles inherit only the values supplied by the caller.
+- Artifact paths identify intended report targets. Confirm actual declared artifact evidence before claiming a durable file.
+- Preserve degraded or insufficient review status. Never turn partial reviewer output into consensus.
+- Release outputs are evidence, not publication authority.
+
+## Supporting Recipes
+
+`git-status`, `git-log`, `changelog-head`, `changelog-section`, `markdown-index`, `package-summary`, and `skill-summary` support the primary workflows. Use one directly only when its narrow deterministic output is the requested result; otherwise start from a primary workflow and avoid rebuilding its composition manually.
+
+## Stop rules
+
+Stop when repository scope, version, validation command, model policy, or write target is ambiguous. Do not edit docs from a `docs-maintenance` plan, publish from release evidence, or send Run Controls from `run-ops`. Hand those operations to their owning protocol after explicit authorization.

package/node_modules/@llblab/pi-actors/skills/project-work/recipes/changelog-head.json

@@ -0,0 +1,11 @@
+{
+  "args": [
+    "file:path",
+    "lines:int"
+  ],
+  "defaults": {
+    "file": "CHANGELOG.md",
+    "lines": "80"
+  },
+  "template": "head -n {lines} {file}"
+}

package/node_modules/@llblab/pi-actors/skills/project-work/recipes/changelog-section.json

@@ -0,0 +1,11 @@
+{
+  "args": [
+    "file:path",
+    "version:string"
+  ],
+  "defaults": {
+    "file": "CHANGELOG.md",
+    "version": "0.7.1"
+  },
+  "template": "{skill_dir}/scripts/project-utils.mjs changelog-section {file} {version}"
+}

package/node_modules/@llblab/pi-actors/skills/project-work/recipes/docs-maintenance.json

@@ -0,0 +1,66 @@
+{
+  "async": true,
+  "imports": {
+    "docs_index": "project-work/markdown-index",
+    "reviewer": "swarm/subagent-review",
+    "planner": "swarm/subagent-plan",
+    "artifact": "artifacts/report"
+  },
+  "args": [
+    "docs_dir:path",
+    "readme:path",
+    "change_summary:string",
+    "artifact_path:path",
+    "model:string",
+    "tools:string"
+  ],
+  "defaults": {
+    "docs_dir": "docs",
+    "readme": "README.md",
+    "change_summary": "Recent project changes require documentation/index consistency review.",
+    "artifact_path": "./docs-maintenance.md",
+    "tools": ""
+  },
+  "template": [
+    {
+      "name": "docs_index",
+      "values": {
+        "directory": "{docs_dir}",
+        "title": "Documentation Index",
+        "include_pattern": "*.md"
+      }
+    },
+    {
+      "name": "reviewer",
+      "values": {
+        "scope": "README {readme}, docs directory {docs_dir}, and docs index output from stdin.",
+        "lens": "documentation maintenance and navigation consistency",
+        "instructions": "Review whether docs navigation, README entrypoints, and changed behavior are coherent. Change summary: {change_summary}",
+        "output_format": "Markdown sections: Missing Links, Stale Docs, Needed Updates, Safe No-op Areas.",
+        "model": "{model}",
+        "tools": "{tools}"
+      }
+    },
+    {
+      "name": "planner",
+      "values": {
+        "goal": "Prepare a bounded docs maintenance plan for: {change_summary}",
+        "scope": "Use docs index and documentation review from stdin. README: {readme}. Docs dir: {docs_dir}.",
+        "constraints": "Prefer smallest documentation changes; do not rewrite stable docs cosmetically.",
+        "output_format": "Markdown sections: Required Edits, Optional Edits, Validation, Stop Conditions.",
+        "model": "{model}",
+        "tools": "{tools}"
+      }
+    },
+    {
+      "name": "artifact",
+      "values": {
+        "input": "Use docs review and maintenance plan from stdin.",
+        "artifact_path": "{artifact_path}",
+        "artifact_format": "Markdown sections: Docs State, Required Updates, Optional Follow-up, Validation.",
+        "model": "{model}",
+        "tools": "{tools}"
+      }
+    }
+  ]
+}

package/node_modules/@llblab/pi-actors/skills/project-work/recipes/git-log.json

@@ -0,0 +1,11 @@
+{
+  "args": [
+    "repo:path",
+    "count:int"
+  ],
+  "defaults": {
+    "repo": ".",
+    "count": "8"
+  },
+  "template": "git -C {repo} log --oneline --decorate -n {count}"
+}

package/node_modules/@llblab/pi-actors/skills/project-work/recipes/git-status.json

@@ -0,0 +1,9 @@
+{
+  "args": [
+    "repo:path"
+  ],
+  "defaults": {
+    "repo": "."
+  },
+  "template": "git -C {repo} status --short --branch"
+}

package/node_modules/@llblab/pi-actors/skills/project-work/recipes/markdown-index.json

@@ -0,0 +1,14 @@
+{
+  "args": [
+    "directory:path",
+    "title:string",
+    "include_pattern:string",
+    "output_format:string"
+  ],
+  "defaults": {
+    "title": "Documentation Index",
+    "include_pattern": "*.md",
+    "output_format": "Markdown bullet list with relative links and one-line descriptions when headings are available."
+  },
+  "template": "find {directory} -maxdepth 1 -name {include_pattern} -type f -printf '%f\\n'"
+}

package/node_modules/@llblab/pi-actors/skills/project-work/recipes/package-summary.json

@@ -0,0 +1,9 @@
+{
+  "args": [
+    "package_json:path"
+  ],
+  "defaults": {
+    "package_json": "package.json"
+  },
+  "template": "{skill_dir}/scripts/project-utils.mjs package-summary {package_json}"
+}

package/node_modules/@llblab/pi-actors/skills/project-work/recipes/release-readiness.json

@@ -0,0 +1,115 @@
+{
+  "async": true,
+  "imports": {
+    "changelog": "project-work/changelog-section",
+    "package": "project-work/package-summary",
+    "skill": "project-work/skill-summary",
+    "validation": "actors/command-validate",
+    "review": "swarm/subagent-review-coordinator",
+    "artifact": "artifacts/report"
+  },
+  "args": [
+    "scope:path",
+    "version:string",
+    "package_json:path",
+    "skill:path",
+    "validation_command:string",
+    "artifact_path:path",
+    "lenses:array",
+    "reviewer_model:string",
+    "verifier_model:string",
+    "merger_model:string",
+    "judge_model:string",
+    "thinking:string",
+    "tools:string",
+    "subagent_ttl_ms:int",
+    "reviewer_concurrency:string",
+    "min_successful_reviewers:int",
+    "merge_policy:string"
+  ],
+  "defaults": {
+    "scope": ".",
+    "version": "0.15.0",
+    "package_json": "{scope}/package.json",
+    "skill": "{scope}/skills/actors/SKILL.md",
+    "validation_command": "npm run validate",
+    "artifact_path": "./release-readiness.md",
+    "lenses": [
+      "release",
+      "package contents",
+      "operator UX",
+      "documentation",
+      "packaged skills"
+    ],
+    "reviewer_model": "{current_model}",
+    "verifier_model": "{current_model}",
+    "merger_model": "{current_model}",
+    "judge_model": "{current_model}",
+    "thinking": "{current_thinking}",
+    "tools": "",
+    "subagent_ttl_ms": 600000,
+    "reviewer_concurrency": "",
+    "min_successful_reviewers": 1,
+    "merge_policy": "Preserve partial reviewer reports, branch status, and degraded confidence markers. If reviewer evidence is below threshold, do not invent consensus."
+  },
+  "template": [
+    {
+      "name": "changelog",
+      "values": {
+        "file": "{scope}/CHANGELOG.md",
+        "version": "{version}"
+      }
+    },
+    {
+      "name": "package",
+      "values": {
+        "package_json": "{package_json}",
+        "repo": "{scope}"
+      }
+    },
+    {
+      "name": "skill",
+      "values": {
+        "skill": "{skill}",
+        "package_json": "{package_json}",
+        "repo": "{scope}"
+      }
+    },
+    {
+      "name": "validation",
+      "values": {
+        "scope": "{scope}",
+        "command": "{validation_command}"
+      }
+    },
+    {
+      "name": "review",
+      "values": {
+        "scope": "Release scope {scope} for version {version}. Use changelog, package summary, packaged skill summary, and validation output from stdin.",
+        "lenses": "{lenses}",
+        "claim": "Version {version} is ready for release or the remaining blockers are explicitly identified.",
+        "reviewer_model": "{reviewer_model}",
+        "verifier_model": "{verifier_model}",
+        "merger_model": "{merger_model}",
+        "judge_model": "{judge_model}",
+        "thinking": "{thinking}",
+        "tools": "{tools}",
+        "subagent_ttl_ms": "{subagent_ttl_ms}",
+        "reviewer_concurrency": "{reviewer_concurrency}",
+        "min_successful_reviewers": "{min_successful_reviewers}",
+        "merge_policy": "{merge_policy}",
+        "output_format": "Markdown sections: Status (complete/degraded/insufficient_data), Release Verdict, Blocking Issues, Package/Docs Risks, Validation Evidence, Publish Notes, Required Follow-up."
+      }
+    },
+    {
+      "name": "artifact",
+      "values": {
+        "input": "Use release readiness review from stdin.",
+        "artifact_path": "{artifact_path}",
+        "artifact_format": "Markdown sections: Release Verdict, Evidence, Risks, Checklist, Next Actions.",
+        "model": "{merger_model}",
+        "tools": "{tools}"
+      }
+    }
+  ]
+}

package/node_modules/@llblab/pi-actors/skills/project-work/recipes/release-summary.json

@@ -0,0 +1,74 @@
+{
+  "async": true,
+  "imports": {
+    "changelog": "project-work/changelog-section",
+    "package": "project-work/package-summary",
+    "skill": "project-work/skill-summary",
+    "validation": "actors/command-validate",
+    "artifact": "artifacts/report"
+  },
+  "args": [
+    "scope:path",
+    "version:string",
+    "package_json:path",
+    "skill:path",
+    "validation_command:string",
+    "artifact_path:path",
+    "risk_notes:string",
+    "model:string",
+    "tools:string"
+  ],
+  "defaults": {
+    "scope": ".",
+    "package_json": "{scope}/package.json",
+    "skill": "{scope}/skills/actors/SKILL.md",
+    "validation_command": "npm run validate",
+    "artifact_path": "./release-summary.md",
+    "risk_notes": "No extra risk notes supplied.",
+    "tools": ""
+  },
+  "artifacts": {
+    "summary": "{artifact_path}"
+  },
+  "template": [
+    {
+      "name": "changelog",
+      "values": {
+        "file": "{scope}/CHANGELOG.md",
+        "version": "{version}"
+      }
+    },
+    {
+      "name": "package",
+      "values": {
+        "package_json": "{package_json}",
+        "repo": "{scope}"
+      }
+    },
+    {
+      "name": "skill",
+      "values": {
+        "skill": "{skill}",
+        "package_json": "{package_json}",
+        "repo": "{scope}"
+      }
+    },
+    {
+      "name": "validation",
+      "values": {
+        "scope": "{scope}",
+        "command": "{validation_command}"
+      }
+    },
+    {
+      "name": "artifact",
+      "values": {
+        "input": "Prepare an evidence-only release summary for version {version}. Use changelog, package summary, packaged skill summary, validation output, and these risk notes: {risk_notes}. Do not propose or perform commit, PR, merge, tag, publish, or any external release side effect.",
+        "artifact_path": "{artifact_path}",
+        "artifact_format": "Markdown sections: Release Summary, Validation Evidence, Package And Skill Evidence, Risks, PR Body Draft, Operator-Gated Next Steps. The PR Body Draft must be suitable for copy/paste and must not claim that release side effects already happened.",
+        "model": "{model}",
+        "tools": "{tools}"
+      }
+    }
+  ]
+}

package/node_modules/@llblab/pi-actors/skills/project-work/recipes/repo-health.json

@@ -0,0 +1,75 @@
+{
+  "async": true,
+  "imports": {
+    "git_status": "project-work/git-status",
+    "git_log": "project-work/git-log",
+    "docs_index": "project-work/markdown-index",
+    "validation": "actors/command-validate",
+    "normalizer": "swarm/subagent-normalize",
+    "artifact": "artifacts/report"
+  },
+  "args": [
+    "repo:path",
+    "docs_dir:path",
+    "validation_command:string",
+    "artifact_path:path",
+    "model:string",
+    "tools:string"
+  ],
+  "defaults": {
+    "repo": ".",
+    "docs_dir": "docs",
+    "validation_command": "npm run validate",
+    "artifact_path": "./repo-health.md",
+    "tools": ""
+  },
+  "template": [
+    {
+      "name": "git_status",
+      "values": {
+        "repo": "{repo}"
+      }
+    },
+    {
+      "name": "git_log",
+      "values": {
+        "repo": "{repo}",
+        "count": "8"
+      }
+    },
+    {
+      "name": "docs_index",
+      "values": {
+        "directory": "{repo}/{docs_dir}",
+        "title": "Documentation Index"
+      }
+    },
+    {
+      "name": "validation",
+      "values": {
+        "scope": "{repo}",
+        "command": "{validation_command}"
+      }
+    },
+    {
+      "name": "normalizer",
+      "values": {
+        "input": "Use repository status, recent log, docs index, and validation output from stdin. Repo: {repo}.",
+        "format": "Markdown sections: Branch State, Recent Commits, Docs Surface, Validation, Risks, Next Actions.",
+        "model": "{model}",
+        "thinking": "off",
+        "tools": "{tools}"
+      }
+    },
+    {
+      "name": "artifact",
+      "values": {
+        "input": "Use normalized repo health output from stdin.",
+        "artifact_path": "{artifact_path}",
+        "artifact_format": "Markdown sections: Branch State, Validation, Risks, Next Actions.",
+        "model": "{model}",
+        "tools": "{tools}"
+      }
+    }
+  ]
+}

package/node_modules/@llblab/pi-actors/skills/project-work/recipes/run-ops.json

@@ -0,0 +1,56 @@
+{
+  "async": true,
+  "imports": {
+    "snapshot": "actors/run-ops-snapshot",
+    "normalizer": "swarm/subagent-normalize",
+    "artifact": "artifacts/report"
+  },
+  "args": [
+    "state_root:path",
+    "run_id:string",
+    "lines:int",
+    "stale_minutes:int",
+    "artifact_path:path",
+    "model:string",
+    "tools:string"
+  ],
+  "defaults": {
+    "state_root": "~/.pi/agent/tmp/pi-actors/runs",
+    "run_id": "music",
+    "lines": "80",
+    "stale_minutes": "60",
+    "artifact_path": "./async-run-ops.md",
+    "tools": ""
+  },
+  "template": [
+    {
+      "name": "snapshot",
+      "values": {
+        "state_root": "{state_root}",
+        "run_id": "{run_id}",
+        "lines": "{lines}",
+        "stale_minutes": "{stale_minutes}"
+      }
+    },
+    {
+      "name": "normalizer",
+      "values": {
+        "input": "Use actor run operations snapshot JSON from stdin. State root: {state_root}. Inspected run: {run_id}.",
+        "format": "Markdown sections: Active Runs, Terminal Runs, Recent Actor Messages, Stale/Risky Runs, Recommended Controls, Follow-up Needed.",
+        "model": "{model}",
+        "thinking": "off",
+        "tools": "{tools}"
+      }
+    },
+    {
+      "name": "artifact",
+      "values": {
+        "input": "Use normalized async-run operations report from stdin.",
+        "artifact_path": "{artifact_path}",
+        "artifact_format": "Markdown sections: Status, Actor Messages, Risks, Recommended Actions.",
+        "model": "{model}",
+        "tools": "{tools}"
+      }
+    }
+  ]
+}

package/node_modules/@llblab/pi-actors/skills/project-work/recipes/skill-summary.json

@@ -0,0 +1,11 @@
+{
+  "args": [
+    "skill:path",
+    "package_json:path"
+  ],
+  "defaults": {
+    "skill": "skills/actors/SKILL.md",
+    "package_json": "package.json"
+  },
+  "template": "{skill_dir}/scripts/project-utils.mjs skill-summary {skill} {package_json}"
+}

package/node_modules/@llblab/pi-actors/skills/project-work/scripts/project-utils.mjs

@@ -0,0 +1,89 @@
+#!/usr/bin/env node
+
+/** Project-work metadata and changelog helpers. */
+
+import { existsSync, readFileSync } from "node:fs";
+import { resolve } from "node:path";
+
+function fail(message) {
+  console.error(message);
+  process.exit(1);
+}
+
+function pathOf(value) {
+  return resolve(String(value).replace(/^~(?=\/|$)/, process.env.HOME ?? "~"));
+}
+
+function readJson(file) {
+  if (!existsSync(file)) return undefined;
+  try {
+    return JSON.parse(readFileSync(file, "utf8"));
+  } catch {
+    return undefined;
+  }
+}
+
+function packageSummary(fileValue = "package.json") {
+  const pkg = readJson(pathOf(fileValue));
+  if (!pkg) fail(`Package JSON not found or invalid: ${fileValue}`);
+  const dependencies = Object.keys(pkg.dependencies ?? {}).sort();
+  const devDependencies = Object.keys(pkg.devDependencies ?? {}).sort();
+  console.log(JSON.stringify({
+    name: pkg.name ?? "",
+    version: pkg.version ?? "",
+    type: pkg.type ?? "",
+    private: Boolean(pkg.private),
+    packageManager: pkg.packageManager ?? "",
+    files: Array.isArray(pkg.files) ? pkg.files : [],
+    bin: pkg.bin ?? null,
+    main: pkg.main ?? "",
+    exports: pkg.exports ?? null,
+    scripts: Object.keys(pkg.scripts ?? {}).sort(),
+    dependencyCount: dependencies.length,
+    devDependencyCount: devDependencies.length,
+    dependencies,
+    devDependencies,
+  }, null, 2));
+}
+
+function skillSummary(skillValue, packageValue = "package.json") {
+  const content = readFileSync(pathOf(skillValue), "utf8");
+  const pkg = readJson(pathOf(packageValue)) ?? {};
+  const frontmatter = content.match(/^---\r?\n([\s\S]*?)\r?\n---/)?.[1] ?? "";
+  const fields = Object.fromEntries(frontmatter.split(/\r?\n/).flatMap((line) => {
+    const match = line.match(/^([A-Za-z_][\w-]*):\s*(.*)$/);
+    return match ? [[match[1], match[2].trim()]] : [];
+  }));
+  const version = fields.version ?? frontmatter.match(/^\s+version:\s*([^\n]+)$/m)?.[1]?.trim() ?? "";
+  const body = content.replace(/^---\r?\n[\s\S]*?\r?\n---\r?\n?/, "");
+  console.log(JSON.stringify({
+    path: skillValue,
+    name: fields.name ?? "",
+    description: fields.description ?? "",
+    version,
+    packageVersion: pkg.version ?? "",
+    versionMatchesPackage: version === pkg.version,
+    frontmatterExtraColonLines: frontmatter.split(/\r?\n/).filter((line) =>
+      /^\w+:\s*\S/.test(line) && (line.match(/:/g) ?? []).length > 1),
+    bodyLineCount: body.split(/\r?\n/).length,
+    headings: body.split(/\r?\n/).filter((line) => /^#{1,6}\s/.test(line)),
+  }, null, 2));
+}
+
+function changelogSection(fileValue, version) {
+  const lines = readFileSync(pathOf(fileValue), "utf8").split(/\r?\n/);
+  const start = lines.findIndex((line) => line.startsWith("## ") && line.includes(version));
+  if (start < 0) fail(`Version section not found: ${version}`);
+  const next = lines.findIndex((line, index) => index > start && line.startsWith("## "));
+  console.log(lines.slice(start, next < 0 ? lines.length : next).join("\n").trimEnd());
+}
+
+const [command, ...args] = process.argv.slice(2);
+try {
+  if (command === "package-summary") packageSummary(args[0]);
+  else if (command === "skill-summary") skillSummary(args[0], args[1]);
+  else if (command === "changelog-section") changelogSection(args[0] ?? "CHANGELOG.md", args[1] ?? "Unreleased");
+  else fail(`Unknown project command: ${command ?? ""}`);
+} catch (error) {
+  fail(error instanceof Error ? error.message : String(error));
+}

package/node_modules/@llblab/pi-actors/skills/recipe-memory/SKILL.md

@@ -0,0 +1,36 @@
+---
+name: recipe-memory
+description: Use only for internal automatic Recipe-memory review, diagnosis, or recovery; do not use for normal Recipe creation, registration, or invocation.
+---
+
+# Recipe Memory
+
+Use this Skill only when automatic draft-memory or active-tool review needs diagnosis or bounded recovery. Normal agents must not spawn, register, specialize, or invoke `recipe-memory/draft-review` or `recipe-memory/tool-review`; they are package-owned internal reviewer components.
+
+## Diagnose first
+
+```text
+inspect target=recipes view=reviews
+inspect target=runtime view=triage
+```
+
+Use the review view to distinguish `draft` and `tool` scope, current phase, failed stage, bounded error, preserved transaction evidence, and the recorded next action. Use runtime triage only when the reviewer Run itself needs Run-level diagnosis. Follow `actors` for generic Inspect, Run, and Control semantics.
+
+## Recovery Controls
+
+Use runtime Controls only when the review evidence names the matching recovery action:
+
+```text
+message target=runtime action=review.retry input={"scope":"draft"}
+message target=runtime action=review.retry input={"scope":"tool"}
+message target=runtime action=review.reset input={"scope":"draft"}
+message target=runtime action=review.reset input={"scope":"tool"}
+```
+
+- `review.retry` resumes the selected failed review scope while preserving authenticated transaction recovery.
+- `review.reset` clears disposable failure state only. It must reject evidence that requires roll-forward recovery.
+- Re-inspect `view=reviews` after one recovery Control; do not loop retries or resets without changed evidence.
+
+## Stop rules
+
+Stop when the phase is healthy or idle, the requested scope does not match the failure, recovery evidence requires deterministic roll-forward, or the reported next action is not retry/reset. Never edit review state, lineage, journals, quarantine, drafts, or persisted Recipes directly. Never bypass automatic review by running its internal Recipes as ordinary capabilities.

package/node_modules/@llblab/pi-actors/skills/recipe-memory/recipes/draft-review.json

@@ -0,0 +1,14 @@
+{
+  "async": true,
+  "args": [
+    "input_path:path",
+    "model:string",
+    "thinking:string"
+  ],
+  "defaults": {
+    "model": "{current_model}",
+    "thinking": "{current_thinking}"
+  },
+  "description": "Read-only background reviewer for one immutable automatic draft-memory batch.",
+  "template": "pi -p --model {model} --thinking {thinking} --no-tools @{input_path} Review the attached immutable pi-actors draft batch. Evaluate every draft independently using launch history, universality, flexibility, parameterization, duplication, safety, and likely future usefulness. Treat draft and sha256 fields as batch-local opaque occurrence/content-group identifiers and return them verbatim. Choose exactly one promote or discard decision per draft; there is no selection quota. Promotion requires a unique absent snake_case target and targetSha256 null. Never return recipe content: the executor derives the exact immutable captured source. Invalid or secret-touching drafts must be discarded. Include assessment with launches, universality, flexibility, futureUsefulness, and safety. Do not mutate, move, register, or delete recipes. End stdout with DRAFT_REVIEW_RESULT on its own line followed by exactly one JSON object with batchId, createdAt, and decisions."
+}