harness-evolver 2.4.0 → 2.5.1

package/bin/install.js

@@ -15,28 +15,19 @@ const VERSION = require("../package.json").version;
 const PLUGIN_ROOT = path.resolve(__dirname, "..");
 const HOME = process.env.HOME || process.env.USERPROFILE;
 
-const MAGENTA = "\x1b[35m";
-const BRIGHT_MAGENTA = "\x1b[95m";
-const GREEN = "\x1b[32m";
+const GREEN = "\x1b[38;2;0;255;136m";
 const YELLOW = "\x1b[33m";
 const RED = "\x1b[31m";
 const DIM = "\x1b[2m";
 const BOLD = "\x1b[1m";
 const RESET = "\x1b[0m";
 
-const LOGO = `
-${BRIGHT_MAGENTA}  ██╗  ██╗ █████╗ ██████╗ ███╗   ██╗███████╗███████╗███████╗
-  ██║  ██║██╔══██╗██╔══██╗████╗  ██║██╔════╝██╔════╝██╔════╝
-  ███████║███████║██████╔╝██╔██╗ ██║█████╗  ███████╗███████╗
-  ██╔══██║██╔══██║██╔══██╗██║╚██╗██║██╔══╝  ╚════██║╚════██║
-  ██║  ██║██║  ██║██║  ██║██║ ╚████║███████╗███████║███████║
-  ╚═╝  ╚═╝╚═╝  ╚═╝╚═╝  ╚═╝╚═╝  ╚═══╝╚══════╝╚══════╝╚══════╝
-  ${MAGENTA}${BOLD}███████╗██╗   ██╗ ██████╗ ██╗    ██╗   ██╗███████╗██████╗
-  ██╔════╝██║   ██║██╔═══██╗██║    ██║   ██║██╔════╝██╔══██╗
-  █████╗  ██║   ██║██║   ██║██║    ██║   ██║█████╗  ██████╔╝
-  ██╔══╝  ╚██╗ ██╔╝██║   ██║██║    ╚██╗ ██╔╝██╔══╝  ██╔══██╗
-  ███████╗ ╚████╔╝ ╚██████╔╝███████╗╚████╔╝ ███████╗██║  ██║
-  ╚══════╝  ╚═══╝   ╚═════╝ ╚══════╝ ╚═══╝  ╚══════╝╚═╝  ╚═╝${RESET}
+const LOGO = `${BOLD}${GREEN}
+  ╦ ╦╔═╗╦═╗╔╗╔╔═╗╔═╗╔═╗  ╔═╗╦  ╦╔═╗╦  ╦  ╦╔═╗╦═╗
+  ╠═╣╠═╣╠╦╝║║║║╣ ╚═╗╚═╗  ║╣ ╚╗╔╝║ ║║  ╚╗╔╝║╣ ╠╦╝
+  ╩ ╩╩ ╩╩╚═╝╚╝╚═╝╚═╝╚═╝  ╚═╝ ╚╝ ╚═╝╩═╝ ╚╝ ╚═╝╩╚═
+${RESET}
+${DIM}${GREEN}  End-to-end harness optimization for AI agents${RESET}
 `;
 
 function ask(rl, question) {
@@ -218,7 +209,7 @@ async function main() {
   console.log();
 
   for (const runtime of selected) {
-    console.log(`  Installing for ${BRIGHT_MAGENTA}${runtime.name}${RESET}\n`);
+    console.log(`  Installing for ${GREEN}${runtime.name}${RESET}\n`);
     cleanupBrokenPluginEntry(runtime.dir);
     installForRuntime(runtime.dir, scope);
     console.log();
@@ -232,7 +223,7 @@ async function main() {
   fs.writeFileSync(versionPath, VERSION);
   console.log(`  ${GREEN}✓${RESET} VERSION ${VERSION}`);
 
-  console.log(`\n  ${GREEN}Done!${RESET} Restart Claude Code, then run ${BRIGHT_MAGENTA}/harness-evolver:init${RESET}\n`);
+  console.log(`\n  ${GREEN}Done!${RESET} Restart Claude Code, then run ${GREEN}/harness-evolver:init${RESET}\n`);
 
   // Optional integrations
   console.log(`  ${YELLOW}Install optional integrations?${RESET}\n`);

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "harness-evolver",
-  "version": "2.4.0",
+  "version": "2.5.1",
   "description": "Meta-Harness-style autonomous harness optimization for Claude Code",
   "author": "Raphael Valdetaro",
   "license": "MIT",

package/skills/init/SKILL.md

@@ -48,30 +48,50 @@ If NO eval exists:
 
 **Tasks** (`tasks/`): If test tasks exist, use them.
 
-If NO tasks exist:
-- Spawn testgen subagent with `subagent_type: "harness-evolver-testgen"`:
-  ```
-  Agent(
-    subagent_type: "harness-evolver-testgen",
-    description: "TestGen: generate test cases for this project",
-    prompt: |
-      <objective>
-      Generate 30 diverse test cases for this project. Write them to tasks/ directory.
-      </objective>
-
-      <files_to_read>
-      - {harness source file path}
-      - {any data files found in the project}
-      </files_to_read>
-
-      <output>
-      Create tasks/ directory with task_001.json through task_030.json.
-      No expected field needed (judge subagent will score outputs).
-      </output>
-  )
-  ```
-- Wait for `## TESTGEN COMPLETE`.
-- Print: "Generated {N} test cases from code analysis."
+If NO tasks exist, generate them. First, identify all relevant source files:
+
+```bash
+find . -name "*.py" -not -path "./.venv/*" -not -path "./.harness-evolver/*" | head -10
+find . -name "*.json" -o -name "*.md" -o -name "*.txt" -o -name "*.yaml" -o -name "*.yml" | grep -v .venv | grep -v .harness-evolver | head -10
+```
+
+Then spawn testgen subagent with CONCRETE file paths (not placeholders):
+
+```
+Agent(
+  subagent_type: "harness-evolver-testgen",
+  description: "TestGen: generate 30 test cases",
+  prompt: |
+    <objective>
+    Generate 30 diverse test cases for this project. Write them to the tasks/ directory
+    in the current working directory.
+    </objective>
+
+    <project_context>
+    This project is at: {absolute path to project root}
+    Entry point: {the harness/agent file you identified, e.g., crew.py or pipeline/moderator.py}
+    Framework: {what you detected — CrewAI, LangGraph, etc.}
+    </project_context>
+
+    <files_to_read>
+    {LIST EVERY .py file and data file you found above — use ABSOLUTE PATHS}
+    Example:
+    - /home/rp/Desktop/test-crewai/crew.py
+    - /home/rp/Desktop/test-crewai/README.md
+    </files_to_read>
+
+    <output>
+    Create directory tasks/ (at project root) with 30 files: task_001.json through task_030.json.
+    Format: {"id": "task_001", "input": "...", "metadata": {"difficulty": "easy|medium|hard", "type": "standard|edge|cross_domain|adversarial"}}
+    No "expected" field needed — the judge subagent will score outputs.
+    Distribution: 40% standard, 20% edge, 20% cross-domain, 20% adversarial.
+    </output>
+)
+```
+
+Wait for `## TESTGEN COMPLETE`. If the subagent fails or returns with no tasks, generate them yourself inline (fallback).
+
+Print: "Generated {N} test cases from code analysis."
 
 ## Phase 3: Run Init