mother-brain 0.4.5 → 0.4.6

package/AGENTS.md

@@ -0,0 +1,63 @@
+# Mother Brain — Agent Instructions
+
+> These rules are MANDATORY for every interaction. They are enforced regardless of whether
+> the Mother Brain skill is explicitly invoked. Violating ANY rule is a critical failure.
+
+## Hard Rules
+
+### 1. Never Leave User in Freeform
+- After completing ANY action (task, release, fix, commit, question, analysis), ALWAYS present a menu with clear options.
+- The user must NEVER see a blank prompt with no guidance on what to do next.
+- If `ask_user` tool is available, use it. Otherwise present numbered options as plain text.
+- If you catch yourself ending a response without a menu → STOP and add one.
+
+### 2. Always Use Menus for Choices
+- EVERY user decision MUST be presented as a menu with selectable options.
+- NEVER ask yes/no questions as plain text.
+- NEVER ask open-ended questions when discrete options exist.
+- When using plain text menus (e.g., Codex CLI), format as:
+  ```
+  1. Option A
+  2. Option B
+  3. Option C
+
+  Reply with the number or option text.
+  ```
+
+### 3. Invoke Child Brain on Freeform Feedback
+- When the user types freeform text instead of selecting a menu option, invoke `$child-brain` IMMEDIATELY.
+- Freeform text = the user has feedback, preferences, or corrections that must be captured.
+- Don't try to handle feedback inline — Child Brain routes learnings to the right place.
+- After Child Brain completes, resume exactly where you were.
+- Trigger keywords: "I prefer", "I like", "actually", "instead", "rather", "maybe", "what about", "stop", "why did you"
+
+### 4. Never Improvise Workflows
+- If Mother Brain skill (`$mother-brain`) is active, follow its Steps section exactly.
+- Do NOT invent menus, skip steps, or make up workflows.
+- If you're unsure what to do next, return to Step 2 (Detect Project State).
+
+### 5. Resume After Skill Invocation
+- After invoking any skill (`$child-brain`, `$skill-creator`, etc.), you MUST return to what you were doing before.
+- NEVER invoke a skill and then stop.
+- Track what step/menu/task you were on, and resume there after the skill completes.
+
+### 6. Show Learning Confirmations
+- When preferences or learnings are recorded, display visible confirmation:
+  - `📘 Project Brain will remember this`
+  - `🧠 Mother Brain will remember this`
+  - `🧙 Elder Brain will remember this`
+- The user should SEE their input being captured — silent learning erodes trust.
+
+### 7. Emoji as Enhancement Only
+- Always include text labels alongside emoji (e.g., "🧠 Mother Brain" not just "🧠").
+- Some runtimes may not render emoji correctly.
+
+## Skills Available
+
+This project uses the Mother Brain framework. The following skills are available:
+
+- **$mother-brain** — Full project management workflow (vision → roadmap → tasks → execution)
+- **$child-brain** — Feedback analysis and learning orchestrator
+- **$skill-creator** — Create new specialized skills
+
+For guided project management, invoke `$mother-brain`.

package/dist/cli.js

@@ -76,6 +76,16 @@ async function init(options = {}) {
     }, { spaces: 2 });
   }
   console.log(chalk.cyan("\n\u2705 Mother Brain initialized!\n"));
+  const agentsFile = path.join(cwd, "AGENTS.md");
+  const sourceAgentsFile = path.join(packageRoot, "AGENTS.md");
+  if (await fs.pathExists(sourceAgentsFile)) {
+    const agentsExists = await fs.pathExists(agentsFile);
+    if (!agentsExists || options.force) {
+      await fs.copy(sourceAgentsFile, agentsFile, { overwrite: true });
+      console.log(chalk.green("Created AGENTS.md (always-active rules for Codex/Copilot)"));
+    }
+  }
+  console.log("");
   console.log("Next steps:");
   console.log(chalk.dim("  1. Commit the new files to your repo"));
   console.log(chalk.dim("  2. Open your AI CLI:"));
@@ -202,6 +212,11 @@ async function update() {
         console.log(chalk2.green(`  \u2713 Updated ${skill}`));
       }
     }
+    const extractedAgentsFile = path2.join(tempDir, "package", "AGENTS.md");
+    const agentsFile = path2.join(cwd, "AGENTS.md");
+    if (await fs2.pathExists(extractedAgentsFile)) {
+      await fs2.copy(extractedAgentsFile, agentsFile, { overwrite: true });
+    }
     await fs2.remove(tempDir);
     await fs2.writeJSON(versionFile, {
       installed: latestVersion,
@@ -747,7 +762,7 @@ async function uninstall(options) {
 // src/cli.ts
 import { exec as exec3 } from "child_process";
 var program = new Command();
-var VERSION = "0.4.5";
+var VERSION = "0.4.6";
 program.name("mother-brain").description("AI-powered project management framework for GitHub Copilot CLI and Codex CLI").version(VERSION);
 program.command("init").description("Initialize Mother Brain in the current project").option("-f, --force", "Overwrite existing skills").action(init);
 program.command("update").description("Update Mother Brain skills to the latest version").action(update);

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "mother-brain",
-  "version": "0.4.5",
+  "version": "0.4.6",
   "description": "AI-powered project management framework for GitHub Copilot CLI",
   "type": "module",
   "main": "dist/index.js",
@@ -29,7 +29,8 @@
   },
   "files": [
     "dist",
-    "skills"
+    "skills",
+    "AGENTS.md"
   ],
   "engines": {
     "node": ">=18"

package/skills/mother-brain/SKILL.md

@@ -249,6 +249,7 @@ Mother Brain transforms high-level visions into executable reality by:
   2. GitHub Release with release notes (use `gh release create` with description)
   3. Update README version badge (if applicable)
   Never publish to npm without also creating a proper GitHub Release with notes.
+- **NEVER END ON FREEFORM**: After completing ANY action (release, fix, learning, commit, task), ALWAYS present a menu with `ask_user` (or numbered plain text in Codex). The user must NEVER see a blank prompt with no guidance. End every action with "What's next?" and concrete options. This applies to releases, commits, fixes, and meta-mode improvements alike.
 - **SESSION STATE IS SOURCE OF TRUTH**: Always read session-state.json AND roadmap.md to determine actual progress. NEVER rely on conversation context alone for task numbering. When determining next task, load roadmap.md and check which tasks have `[ ]` vs `[x]`. Wrong task numbers destroy user trust—always verify against files, not memory.
 - **ROADMAP CHECKBOX UPDATE (MANDATORY)**: After EVERY task is marked complete, IMMEDIATELY update roadmap.md to check off that task's checkbox (`[ ]` → `[x]`). This is NOT optional and NOT deferred. Stale checkboxes are a critical failure—roadmap must always reflect reality. Use `edit` tool to update the specific task line in roadmap.md right after user confirms task completion.
 - **END-TO-END WALKTHROUGH FOR NEW INTEGRATIONS**: After implementing a new integration or feature (especially cross-tool like CLI→Codex, API→frontend), proactively walk the user through how to use it end-to-end BEFORE marking the task complete. Don't assume the user knows the invocation syntax, required steps, or expected workflow. Show concrete commands and expected output.