claude-turing 4.3.0 → 4.5.0

package/bin/cli.js

@@ -1,9 +1,25 @@
 #!/usr/bin/env node
 import { createRequire } from "module";
+import { realpathSync } from "fs";
+import { fileURLToPath } from "url";
 const require = createRequire(import.meta.url);
 const { Command } = require("commander");
 const pkg = require("../package.json");
 
+export function buildInitArgs(name, dir) {
+  return [name, dir].filter(Boolean);
+}
+
+function isDirectRun() {
+  if (!process.argv[1]) return false;
+
+  try {
+    return realpathSync(fileURLToPath(import.meta.url)) === realpathSync(process.argv[1]);
+  } catch {
+    return false;
+  }
+}
+
 const program = new Command();
 
 program
@@ -34,13 +50,16 @@ program
   .command("init [name] [dir]")
   .description("Scaffold ML project (CLI mode, non-Claude-Code usage)")
   .action(async (name, dir) => {
-    const { execSync } = await import("child_process");
+    const { spawnSync } = await import("child_process");
     const { dirname, join } = await import("path");
     const { fileURLToPath } = await import("url");
     const __dirname = dirname(fileURLToPath(import.meta.url));
     const script = join(__dirname, "turing-init.sh");
-    const args = [name, dir].filter(Boolean).join(" ");
-    execSync(`bash "${script}" ${args}`, { stdio: "inherit" });
+    const args = buildInitArgs(name, dir);
+    const result = spawnSync("bash", [script, ...args], { stdio: "inherit" });
+    process.exit(result.status ?? 1);
   });
 
-program.parse();
+if (isDirectRun()) {
+  program.parse();
+}

package/commands/doctor.md

@@ -0,0 +1,31 @@
+---
+name: doctor
+description: Harness self-diagnosis — check environment, project, resources, and git state. Auto-fix common issues.
+disable-model-invocation: true
+argument-hint: "[--fix] [--verbose]"
+allowed-tools: Read, Bash(*), Grep, Glob
+---
+
+Is Turing healthy? Check everything and get a score.
+
+## Steps
+1. `source .venv/bin/activate`
+2. `python scripts/harness_doctor.py $ARGUMENTS`
+3. **Saved:** `experiments/doctor/`
+
+## Checks
+- **Environment:** Python version, venv status
+- **Dependencies:** all required packages importable
+- **Config:** config.yaml valid with required fields
+- **Experiment log:** JSONL integrity, corrupt line detection
+- **Scripts:** train.py, prepare.py, evaluate.py exist and parse
+- **Disk space:** warn if <1GB free
+- **Git state:** uncommitted changes to critical files
+- **Claude hooks:** `.claude/settings.local.json` hook group schema; `--fix` migrates legacy bare command hooks
+
+## Examples
+```
+/turing:doctor
+/turing:doctor --fix
+/turing:doctor --verbose --json
+```

package/commands/init.md

@@ -37,18 +37,36 @@ python3 <templates_dir>/scripts/scaffold.py \
 The scaffold script handles everything in a single atomic operation:
 - Copies all template files with placeholder substitution
 - Creates data/, experiments/, models/ directories
-- Sets up agent memory at `.claude/agent-memory/ml-researcher/MEMORY.md`
+- Sets up agent memory at `.claude/agent-memory/ml-researcher-{project_name}/MEMORY.md`
 - Configures Claude Code hooks in `.claude/settings.local.json`
 - Creates Python virtual environment and installs requirements
 - Verifies all placeholders were replaced (fails loudly if any remain)
 
 ## Locating Templates
 
-Find the templates directory using Glob:
+Use the installed command-pack templates directory first:
+```
+.claude/commands/turing/templates/
+~/.claude/commands/turing/templates/
+```
+Then fall back to plugin or npm locations:
 ```
 ~/.claude/plugins/*/templates/
+node_modules/claude-turing/templates/
+```
+
+Example command:
+
+```bash
+python3 ~/.claude/commands/turing/templates/scripts/scaffold.py \
+    --project-name "<project_name>" \
+    --target-metric "<target_metric>" \
+    --metric-direction "<metric_direction>" \
+    --task-description "<task_description>" \
+    --ml-dir "<ml_dir>" \
+    --data-source "<data_source>" \
+    --templates-dir ~/.claude/commands/turing/templates
 ```
-Or check if installed via npm by looking for `node_modules/claude-turing/templates/`.
 
 ## After Scaffolding
 

package/commands/plan.md

@@ -0,0 +1,27 @@
+---
+name: plan
+description: Research planning assistant — design a strategic experiment campaign with budget-aware ROI allocation.
+disable-model-invocation: true
+argument-hint: "[--budget 20] [--goal \"maximize F1 for production\"]"
+allowed-tools: Read, Bash(*), Grep, Glob
+---
+
+Design the next N experiments strategically, not randomly. Allocates budget by expected ROI.
+
+## Steps
+1. `source .venv/bin/activate`
+2. `python scripts/research_planner.py $ARGUMENTS`
+3. **Saved:** `experiments/plans/`
+
+## How it works
+- Analyzes experiment history to compute per-family ROI
+- Adjusts strategy priorities based on project state and goal
+- Allocates budget across: feature engineering, model search, ensemble, calibration, verification
+- Generates phased plan with specific experiment descriptions
+
+## Examples
+```
+/turing:plan --budget 20
+/turing:plan --budget 10 --goal "maximize F1 for production deployment"
+/turing:plan --budget 30 --json
+```

package/commands/postmortem.md

@@ -0,0 +1,28 @@
+---
+name: postmortem
+description: Failure postmortem — diagnose why experiments stopped improving and get actionable next steps.
+disable-model-invocation: true
+argument-hint: "[--window 10] [--auto-trigger 5]"
+allowed-tools: Read, Bash(*), Grep, Glob
+---
+
+When experiments stop improving, find out why. Diagnoses search space exhaustion, config errors, data issues, metric ceilings, and noise floors.
+
+## Steps
+1. `source .venv/bin/activate`
+2. `python scripts/failure_postmortem.py $ARGUMENTS`
+3. **Saved:** `experiments/postmortems/`
+
+## Diagnosis categories
+- **Search space exhaustion:** micro-tuning params that don't matter
+- **Systematic config error:** all experiments share a bad common config
+- **Data issue:** all model types fail similarly
+- **Metric ceiling:** near theoretical maximum
+- **Noise floor:** improvements within seed variance
+
+## Examples
+```
+/turing:postmortem
+/turing:postmortem --window 15
+/turing:postmortem --json
+```

package/commands/turing.md

@@ -80,6 +80,9 @@ You are the Turing ML research router. Detect the user's intent and route to the
 | "simulate", "predict outcome", "pre-filter", "which configs will work", "forecast" | `/turing:simulate` | Predict |
 | "update", "incremental", "new data", "add data", "fine-tune existing", "partial update" | `/turing:update` | Update |
 | "registry", "promote", "demote", "staging", "production", "which model is deployed", "model lifecycle" | `/turing:registry` | Govern |
+| "postmortem", "why failing", "failure streak", "why no improvement", "what went wrong" | `/turing:postmortem` | Diagnose |
+| "doctor", "health check", "is it broken", "diagnose harness", "self-check" | `/turing:doctor` | Check |
+| "plan", "research plan", "campaign", "what next", "allocate budget", "strategic plan" | `/turing:plan` | Plan |
 
 ## Sub-commands
 
@@ -156,6 +159,9 @@ You are the Turing ML research router. Detect the user's intent and route to the
 | `/turing:simulate [--configs] [--top-k]` | Experiment outcome prediction: pre-filter configs using surrogate model, save budget | (inline) |
 | `/turing:update <exp-id> --new-data <path>` | Incremental model update: add new data without full retraining, forgetting detection | (inline) |
 | `/turing:registry [list\|register\|promote\|demote\|history]` | Model registry: stage lifecycle (candidate → staging → production) with promotion gates | (inline) |
+| `/turing:postmortem [--window N]` | Failure postmortem: diagnose why experiments stopped improving (exhaustion, config error, data issue, ceiling, noise) | (inline) |
+| `/turing:doctor [--fix]` | Harness self-diagnosis: environment, dependencies, config, log integrity, scripts, disk, git state, Claude hooks | (inline) |
+| `/turing:plan [--budget N] [--goal]` | Research planning assistant: strategic campaign design with budget-aware ROI allocation | (inline) |
 
 ## Proactive Detection
 

package/config/defaults.yaml

@@ -35,6 +35,8 @@ placeholders:
   ML_DIR: "Directory for ML files relative to project root"
   DATA_SOURCE: "Path to training data file"
   METRIC_DIRECTION: "lower or higher — which direction is better"
+  LOWER_IS_BETTER: "Boolean derived from METRIC_DIRECTION for generated config.yaml"
+  MEMORY_DIR_NAME: "Filesystem-safe project name used for scoped agent memory"
 
 # Agent configuration
 agents:

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "claude-turing",
-  "version": "4.3.0",
+  "version": "4.5.0",
   "type": "module",
   "description": "Autonomous ML research harness for Claude Code. The autoresearch loop as a formal protocol — iteratively trains, evaluates, and improves ML models with structured experiment tracking, convergence detection, immutable evaluation infrastructure, and safety guardrails.",
   "bin": {
@@ -34,15 +34,15 @@
     "ai-agents"
   ],
   "author": {
-    "name": "pragnition"
+    "name": "Prannaya Gupta"
   },
-  "homepage": "https://github.com/pragnition/turing",
+  "homepage": "https://github.com/ThePyProgrammer/turing",
   "repository": {
     "type": "git",
-    "url": "git+https://github.com/pragnition/turing.git"
+    "url": "git+https://github.com/ThePyProgrammer/turing.git"
   },
   "bugs": {
-    "url": "https://github.com/pragnition/turing/issues"
+    "url": "https://github.com/ThePyProgrammer/turing/issues"
   },
   "license": "MIT",
   "engines": {

package/src/install.js

@@ -9,7 +9,7 @@
  *   node src/install.js [--global] [--project]
  */
 
-import { readdir, copyFile, mkdir } from "fs/promises";
+import { readdir, copyFile, mkdir, cp } from "fs/promises";
 import { join, dirname } from "path";
 import { fileURLToPath } from "url";
 import { getTargetPaths } from "./paths.js";
@@ -38,6 +38,7 @@ const SUB_COMMANDS = [
   "onboard", "share", "review",
   "whatif", "counterfactual", "simulate",
   "update", "registry",
+  "postmortem", "doctor", "plan",
 ];
 
 export async function install(opts = {}) {
@@ -49,7 +50,7 @@ export async function install(opts = {}) {
   console.log("");
 
   // Create directories for each sub-command + agents + config
-  for (const subDir of ["", "agents", "config", "rules", ...SUB_COMMANDS]) {
+  for (const subDir of ["", "agents", "config", "rules", "templates", ...SUB_COMMANDS]) {
     await mkdir(join(paths.commands, subDir), { recursive: true });
   }
 
@@ -102,6 +103,21 @@ export async function install(opts = {}) {
   }
   console.log(`  ${CONFIG_FILES.length} config files installed`);
 
+  // Copy templates used by /turing:init
+  await cp(
+    join(PLUGIN_ROOT, "templates"),
+    join(paths.commands, "templates"),
+    {
+      recursive: true,
+      force: true,
+      filter: (src) =>
+        !src.includes("__pycache__") &&
+        !src.includes(".pytest_cache") &&
+        !src.endsWith(".pyc"),
+    },
+  );
+  console.log("  Templates installed");
+
   // Update CLAUDE.md
   await updateClaudeMd(paths.claudeMd);
   console.log("  CLAUDE.md updated");

package/src/verify.js

@@ -8,10 +8,14 @@
  *   node src/verify.js [--scope global|project]
  */
 
-import { access } from "fs/promises";
-import { join } from "path";
+import { access, readdir } from "fs/promises";
+import { dirname, join } from "path";
+import { fileURLToPath } from "url";
 import { getTargetPaths } from "./paths.js";
 
+const __dirname = dirname(fileURLToPath(import.meta.url));
+const PLUGIN_ROOT = join(__dirname, "..");
+
 const EXPECTED_COMMANDS = [
   "SKILL.md",
   "init/SKILL.md",
@@ -85,6 +89,9 @@ const EXPECTED_COMMANDS = [
   "simulate/SKILL.md",
   "update/SKILL.md",
   "registry/SKILL.md",
+  "postmortem/SKILL.md",
+  "doctor/SKILL.md",
+  "plan/SKILL.md",
 ];
 
 const EXPECTED_AGENTS = ["ml-researcher.md", "ml-evaluator.md"];
@@ -97,6 +104,27 @@ const EXPECTED_CONFIG = [
   "watch_alerts.yaml",
 ];
 
+async function templateFiles(root, relativeDir = "templates") {
+  const dir = join(root, relativeDir);
+  const entries = await readdir(dir, { withFileTypes: true });
+  const files = [];
+
+  for (const entry of entries) {
+    if (entry.name === "__pycache__" || entry.name === ".pytest_cache") {
+      continue;
+    }
+
+    const relativePath = `${relativeDir}/${entry.name}`;
+    if (entry.isDirectory()) {
+      files.push(...await templateFiles(root, relativePath));
+    } else if (!entry.name.endsWith(".pyc")) {
+      files.push(relativePath);
+    }
+  }
+
+  return files;
+}
+
 async function fileExists(path) {
   try {
     await access(path);
@@ -108,7 +136,9 @@ async function fileExists(path) {
 
 export async function verify(opts = {}) {
   const scopes = opts.scope ? [opts.scope] : ["global", "project"];
+  const expectedTemplates = await templateFiles(PLUGIN_ROOT);
   let found = false;
+  let totalMissing = 0;
 
   for (const scope of scopes) {
     const paths = getTargetPaths(scope);
@@ -141,10 +171,18 @@ export async function verify(opts = {}) {
       if (!ok) missing++;
     }
 
+    console.log("\nTemplates:");
+    for (const template of expectedTemplates) {
+      const ok = await fileExists(join(paths.commands, template));
+      console.log(`  ${ok ? "✓" : "✗"} commands/${template}`);
+      if (!ok) missing++;
+    }
+
     // Check CLAUDE.md
     const claudeOk = await fileExists(paths.claudeMd);
     console.log(`\n  ${claudeOk ? "✓" : "✗"} CLAUDE.md`);
 
+    totalMissing += missing;
     console.log(
       `\n  ${missing === 0 ? "✓ Installation complete" : `✗ ${missing} files missing — run claude-turing install`}\n`,
     );
@@ -152,6 +190,11 @@ export async function verify(opts = {}) {
 
   if (!found) {
     console.log("\n✗ turing not found. Run: claude-turing install\n");
+    totalMissing++;
+  }
+
+  if (totalMissing > 0) {
+    process.exitCode = 1;
   }
 }
 

package/templates/README.md

@@ -4,7 +4,7 @@
 
 ## Overview
 
-This pipeline uses the [Turing](https://github.com/pragnition/turing) autoresearch pattern — an AI agent iteratively trains, evaluates, and improves models by modifying `train.py` while the evaluation infrastructure (`prepare.py`, `evaluate.py`) remains immutable.
+This pipeline uses the [Turing](https://github.com/ThePyProgrammer/turing) autoresearch pattern — an AI agent iteratively trains, evaluates, and improves models by modifying `train.py` while the evaluation infrastructure (`prepare.py`, `evaluate.py`) remains immutable.
 
 **Primary metric:** {{TARGET_METRIC}} ({{METRIC_DIRECTION}} is better)
 

package/templates/config.yaml

@@ -17,7 +17,7 @@ evaluation:
   metrics: ["{{TARGET_METRIC}}", "f1_weighted", "accuracy"]
   # Set to true for metrics where lower is better (mae, mse, rmse, loss)
   # Set to false for metrics where higher is better (accuracy, f1, auc)
-  lower_is_better: false  # {{METRIC_DIRECTION}} -- change to true if lower is better
+  lower_is_better: {{LOWER_IS_BETTER}}
 
   # Multi-seed configuration (Phase 10.1: /turing:seed)
   # Seeds used for seed studies — diverse values for good coverage

package/templates/program.md

@@ -43,7 +43,7 @@ git checkout -b exp/NNN-description
 
 ## Memory
 
-Read `.claude/agent-memory/ml-researcher/MEMORY.md` at the start of each session.
+Read `.claude/agent-memory/ml-researcher-{{MEMORY_DIR_NAME}}/MEMORY.md` at the start of each session.
 Update it after each experiment with:
 - Best result (if improved)
 - What was tried and why