pi-hermes-memory 0.7.6 → 0.7.8

package/README.md

@@ -69,7 +69,7 @@ The extension manages three types of knowledge:
 |---|---|---|---|
 | **Memory** (MEMORY.md) | Facts — env details, project conventions, tool quirks | 5,000 chars max | Searchable by default |
 | **User Profile** (USER.md) | Who you are — name, preferences, communication style | 5,000 chars max | Searchable by default |
-| **Skills** (skills/*.md) | Procedures — *how* to do something, reusable across sessions | Unlimited | Available through skill tool |
+| **Skills** (Pi-native `SKILL.md`) | Procedures — *how* to do something, reusable across sessions | Unlimited | Discoverable by Pi + manageable via the skill tool |
 
 ![Memory + Skills Architecture](docs/images/memory-architecture.svg)
 
@@ -119,7 +119,7 @@ System Prompt
 └─────────────────────────────────────────┘
 ```
 
-Set `"memoryPolicyStyle"` to `"full"`, `"compact"`, `"custom"`, or `"none"` to choose policy verbosity while keeping policy-only mode. Set `"memoryMode": "legacy-inject"` to restore the old behavior that injects MEMORY.md, USER.md, project memory, recent failures, and the skill index into the prompt.
+Set `"memoryPolicyStyle"` to `"full"`, `"compact"`, `"custom"`, or `"none"` to choose policy verbosity while keeping policy-only mode. Set `"memoryMode": "legacy-inject"` to restore the old behavior that injects MEMORY.md, USER.md, project memory, and recent failures into the prompt.
 
 ## Failure Memory
 
@@ -177,13 +177,29 @@ The agent also gets a `skill` tool for saving reusable procedures:
 
 | Action | What it does |
 |---|---|
-| `create` | Save a new skill (name, description, step-by-step body) |
-| `view` | Read a skill's full content, or list all skills if no name given |
-| `patch` | Update one section of an existing skill (e.g., just the Procedure section) |
-| `edit` | Replace the description and/or full body of a skill |
-| `delete` | Remove a skill |
+| `create` | Save a new skill (name, description, step-by-step body, optional `scope`) |
+| `view` | Read a skill's full content by `skill_id`, or list all skills if no id is given |
+| `patch` | Update one section of an existing skill by `skill_id` |
+| `edit` | Replace the description and/or full body of a skill by `skill_id` |
+| `delete` | Remove a skill by `skill_id` |
 
-Skills are stored as `SKILL.md` files in `~/.pi/agent/memory/skills/`. Each has a structured body:
+Skills are stored in Pi-native locations:
+
+- Global skills: `~/.pi/agent/skills/<slug>/SKILL.md`
+- Project skills: `~/.pi/agent/projects-memory/<project>/skills/<slug>/SKILL.md`
+
+The extension classifies new skills automatically:
+
+- `global` for transferable procedures
+- `project` for repo-specific workflows tied to local paths, scripts, architecture, deploy steps, or conventions
+
+Global skill creation also has duplicate/similarity guards:
+
+- exact slug match → blocked (update existing via `patch`/`edit`)
+- near-name + high description similarity → blocked as similar (enhance existing)
+- near-name + low description similarity → blocked as name collision (rename to a clearer distinct skill name)
+
+Each skill uses a structured `SKILL.md` body:
 
 ```markdown
 ---
@@ -209,13 +225,23 @@ When you see TypeScript compilation errors, especially in monorepo setups.
 Run `tsc --noEmit` and confirm zero errors.
 ```
 
+### Project Skill Discovery (`resources_discover`)
+
+Project-scoped skills are loaded via Pi's `resources_discover` hook.
+
+On discovery, the extension returns the active project's skills directory as a skill path:
+
+- `~/.pi/agent/projects-memory/<project>/skills/`
+
+This lets Pi discover project skills as native skills without copying them into the global skills folder.
+
 ### Memory vs User Profile vs Skills
 
 | Store | File | What goes here | Limit |
 |---|---|---|---|
 | **memory** | `MEMORY.md` | Agent's notes — env facts, project conventions, tool quirks, lessons learned | 5,000 chars |
 | **user** | `USER.md` | User profile — name, preferences, communication style, habits | 5,000 chars |
-| **skills** | `skills/*.md` | Procedures — *how* to debug, deploy, test, or fix something | Unlimited |
+| **skills** | `~/.pi/agent/skills/<slug>/SKILL.md` or `projects-memory/<project>/skills/<slug>/SKILL.md` | Procedures — *how* to debug, deploy, test, or fix something | Unlimited |
 | **extended** | `sessions.db` | Searchable memories beyond the core limit | Unlimited |
 | **sessions** | `sessions.db` | Past conversation history (searchable via FTS5) | Unlimited |
 
@@ -295,13 +321,13 @@ This means skills build up naturally over time without you having to ask.
 | Command | What it does |
 |---|---|
 | `/memory-insights` | Shows everything stored in memory and user profile |
-| `/memory-skills` | Lists all agent-created skills |
+| `/memory-skills` | Opens an interactive skills manager for search, multi-select, move, and delete |
 | `/memory-consolidate` | Manually trigger memory consolidation to free space |
 | `/memory-interview` | Answer a few questions to pre-fill your user profile |
 | `/memory-switch-project` | List all project memories and their entry counts |
 | `/memory-index-sessions` | Import past Pi sessions into the search database |
 | `/memory-sync-markdown` | Backfill Markdown memories into the SQLite search store |
-| `/memory-preview-context` | Preview the memory policy or legacy memory/skill blocks appended to the system prompt |
+| `/memory-preview-context` | Preview the memory policy or legacy memory blocks appended to the system prompt |
 | `/learn-memory-tool` | Skill that teaches users how to use the memory system |
 
 ### `/memory-insights` Output
@@ -324,21 +350,34 @@ This means skills build up naturally over time without you having to ask.
 3. codes primarily in TypeScript
 ```
 
-### `/memory-skills` Output
-
-```
-╔══════════════════════════════════════════════╗
-║            🧠 Procedural Skills             ║
-╚══════════════════════════════════════════════╝
-
-📄 debug-typescript-errors
-   Step-by-step approach to debugging TS errors in monorepos
-   file: debug-typescript-errors.md
-
-📄 deploy-checklist
-   Pre-deploy verification steps for this project
-   file: deploy-checklist.md
-```
+### `/memory-skills` Manager
+
+`/memory-skills` now opens an interactive TUI modal for skill management.
+
+Features:
+- fuzzy search by skill name
+- single-list view with scope badges (`[G]` global, `[P]` project)
+- multi-select with spacebar
+- batch move to global or current project
+- batch delete with one confirmation
+- inline action summaries for partial success/conflicts
+
+Keybindings:
+- `↑` / `↓` — move focus
+- `space` — toggle selection
+- `/` — focus search
+- `tab` — switch between search and list
+- `g` — move selected skills to global
+- `p` — move selected skills to project
+- `d` — delete selected skills
+- `a` — select all filtered skills
+- `n` — clear selection
+- `esc` — close the modal
+
+Move behavior:
+- moves are **conflict-safe**
+- if the destination already contains the same slug, the conflicting skill stays in place
+- batch moves use partial-success semantics: non-conflicting skills move, blocked skills are reported in the summary
 
 ## Configuration
 
@@ -373,7 +412,7 @@ Create `~/.pi/agent/hermes-memory-config.json`:
 
 | Setting | Default | Description |
 |---|---|---|
-| `memoryMode` | `policy-only` | Prompt behavior: `policy-only` injects only memory policy; `legacy-inject` restores full memory/skill prompt injection |
+| `memoryMode` | `policy-only` | Prompt behavior: `policy-only` injects only memory policy; `legacy-inject` restores full memory prompt injection |
 | `memoryPolicyStyle` | `full` | Policy text used in `policy-only` mode: `full` preserves the default v0.7 policy; `compact` uses shorter built-in guidance; `custom` uses `memoryPolicyCustomText`; `none` injects no policy text |
 | `memoryPolicyCustomText` | unset | Custom policy text used when `memoryPolicyStyle` is `custom`; blank or missing text falls back to `compact` |
 | `memoryCharLimit` | `5000` | Max characters in MEMORY.md |
@@ -405,23 +444,29 @@ Create `~/.pi/agent/hermes-memory-config.json`:
 
 ```
 ~/.pi/agent/
+├── skills/                ← Global Pi-native skills
+│   ├── debug-typescript-errors/
+│   │   └── SKILL.md
+│   └── testing-checklist/
+│       └── SKILL.md
 ├── projects-memory/       ← ALL project-scoped memories (one subfolder per project)
 │   ├── my-project/
-│   │   └── MEMORY.md
+│   │   ├── MEMORY.md
+│   │   └── skills/
+│   │       └── deploy-checklist/
+│   │           └── SKILL.md
 │   └── another-project/
 │       └── MEMORY.md
 ├── memory/                ← Global memory
 │   ├── MEMORY.md          ← Agent's personal notes (env facts, patterns, lessons)
 │   ├── USER.md            ← User profile (name, preferences, habits)
 │   ├── sessions.db        ← SQLite database (session history + extended memory)
-│   └── skills/
-│       ├── debug-typescript-errors.md
-│       └── deploy-checklist.md
+│   └── .skills-migrated-to-pi-native
 ├── hermes-memory-config.json
 └── ...
 ```
 
-These are plain markdown files. You can read and edit them directly if you want to curate what the agent remembers. Memory entries are separated by `§` (section sign). Skills use standard SKILL.md format with frontmatter.
+These are plain markdown files. You can read and edit them directly if you want to curate what the agent remembers. Memory entries are separated by `§` (section sign). Skills use Pi-compatible `SKILL.md` files with frontmatter.
 
 If you are upgrading from a version that stored project memory directly at `~/.pi/agent/<project>/MEMORY.md`, the extension copies or merges those entries into `~/.pi/agent/projects-memory/<project>/MEMORY.md` on startup. The old folders are left in place as a backup.
 
@@ -435,7 +480,9 @@ The `sessions.db` SQLite database stores session history and extended memory ent
 - **Older Markdown memories may need backfill**: If you saved memories before the SQLite mirror existed or search looks stale, run `/memory-sync-markdown`.
 - **Core memory limits still apply**: SQLite search mirroring does not bypass the 5,000-char core Markdown limit. If consolidation cannot free space, the write fails instead of becoming SQLite-only memory invisibly.
 - **System prompts are invisible**: Pi's TUI does not display the system prompt. Use `/memory-preview-context` to inspect whether policy-only or legacy memory injection is active.
-- **Skills are agent-generated**: Skills are created by the agent based on its experience. They may not always be perfectly structured. You can edit or delete them in `~/.pi/agent/memory/skills/`.
+- **Project skill visibility depends on Pi discovery cycles**: project skills are exposed through `resources_discover` using the active project's `skills/` path. If a moved or newly created project skill doesn't show up immediately in a running session, trigger a reload/new session so Pi refreshes discovered resources.
+- **Project move requires active project context**: in `/memory-skills`, the `p` hotkey is disabled when Pi is not currently in a detected project directory.
+- **Skills are agent-generated**: Skills are created by the agent based on its experience. They may not always be perfectly structured. You can move, delete, or still edit them directly in `~/.pi/agent/skills/` or the active project's `skills/` folder.
 
 ## Architecture
 

package/docs/PUBLISHING.md

@@ -45,7 +45,7 @@ Target state:
     "image": "https://raw.githubusercontent.com/chandra447/pi-hermes-memory/main/docs/assets/hermes-memory-preview.png"
   },
   "peerDependencies": {
-    "@mariozechner/pi-coding-agent": "*"
+    "@earendil-works/pi-coding-agent": "*"
   }
 }
 ```

package/docs/ROADMAP.md

@@ -131,7 +131,7 @@ Hermes creates skills after complex tasks (5+ tool calls). Skills are SKILL.md f
 **Key insight**: Pi already has a skill system. Our skill tool should write SKILL.md files that are compatible with Pi's skill discovery. This means our skills are immediately usable as Pi slash commands — no separate ecosystem needed.
 
 - [ ] `skill` tool — register via `pi.registerTool()` with actions: `create`, `patch`, `edit`, `delete`
-- [ ] Skill storage in `~/.pi/agent/memory/skills/` (not `~/.pi/agent/skills/` — avoid conflicting with user's own skills)
+- [x] Skill storage split by scope: global skills in `~/.pi/agent/skills/<slug>/SKILL.md`, project skills in `~/.pi/agent/projects-memory/<project>/skills/<slug>/SKILL.md` (discovered via `resources_discover`)
 - [ ] SKILL.md format — compatible with Pi's SKILL.md spec (frontmatter + markdown body)
 - [ ] Progressive disclosure — skill index (name + description only) injected into system prompt, full content loaded on demand via `skill_view` action
 - [ ] Auto-trigger after complex tasks — track tool calls per turn, trigger skill extraction at 5+ tool calls

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "pi-hermes-memory",
-  "version": "0.7.6",
+  "version": "0.7.8",
   "description": "🧠 Persistent memory + 🔍 session search + 🛡️ secret scanning for Pi. Token-aware policy-only memory by default, SQLite FTS5 search, auto-consolidation, procedural skills. 368 tests. Ported from Hermes agent.",
   "type": "module",
   "main": "src/index.ts",
@@ -45,17 +45,18 @@
     "url": "https://github.com/chandra447/pi-hermes-memory"
   },
   "peerDependencies": {
-    "@mariozechner/pi-coding-agent": ">=0.70.0"
+    "@earendil-works/pi-coding-agent": ">=0.74.0"
   },
   "devDependencies": {
-    "@mariozechner/pi-ai": "^0.70.0",
-    "@mariozechner/pi-coding-agent": "^0.70.0",
+    "@earendil-works/pi-ai": "^0.74.0",
+    "@earendil-works/pi-coding-agent": "^0.74.0",
     "@types/better-sqlite3": "^7.6.13",
     "tsx": "^4.21.0",
     "typebox": "^1.1.33",
     "typescript": "^6.0.3"
   },
   "dependencies": {
+    "@earendil-works/pi-tui": "^0.74.0",
     "better-sqlite3": "^12.9.0"
   }
 }

package/src/constants.ts

@@ -134,7 +134,7 @@ export const COMBINED_REVIEW_PROMPT = `Review the conversation above and conside
 
 For failures, include: what was tried, why it failed, what error occurred, and what worked instead.
 
-**Skills**: Was a complex, non-trivial approach used to complete a task — one that required trial and error, multiple tool calls, or changing course? If so, save a reusable procedure using the skill tool with action 'create'. Include: when to use it, step-by-step procedure, pitfalls to avoid, and how to verify success. If a related skill already exists, use action 'patch' to update it instead of creating a duplicate.
+**Skills**: Was a complex, non-trivial approach used to complete a task — one that required trial and error, multiple tool calls, or changing course? If so, save a reusable procedure using the skill tool with action 'create'. Always pass scope explicitly on create (scope='global' or scope='project'). Choose scope='global' for transferable procedures and scope='project' when the workflow depends on this repo's paths, scripts, architecture, deploy steps, or conventions. Include: when to use it, step-by-step procedure, pitfalls to avoid, and how to verify success. If a related skill already exists, use action 'patch' with its skill_id instead of creating a duplicate.
 
 Only act if there's something genuinely worth saving. If nothing stands out, just say 'Nothing to save.' and stop.`;
 
@@ -221,13 +221,17 @@ Priority:
 Use the memory tool to save. If this contradicts an existing entry, use 'replace' to update it.`;
 
 // ─── Skill tool description ───
-export const SKILL_TOOL_DESCRIPTION = `Save reusable procedures and patterns as skills that survive across sessions. Skills are procedural memory — they capture HOW to do something, not just what happened.
+export const SKILL_TOOL_DESCRIPTION = `Save reusable procedures and patterns as Pi-native skills that survive across sessions. Skills are procedural memory — they capture HOW to do something, not just what happened.
 
 WHEN TO CREATE A SKILL:
 - After completing a complex task that required trial and error or multiple tool calls
 - When you discover a non-obvious approach that could be reused
 - When the user teaches you a specific workflow or procedure
 
+SCOPE:
+- 'global': transferable procedures that can be reused across repositories
+- 'project': procedures tied to this repo's paths, scripts, architecture, deploy flow, or conventions
+
 WHEN TO UPDATE A SKILL (use 'patch'):
 - You discover a better approach for an existing skill
 - A pitfall or edge case not covered by the skill
@@ -238,7 +242,7 @@ SKILL FORMAT:
 - description: one-line summary of when to use it
 - body: structured with sections — ## When to Use, ## Procedure, ## Pitfalls, ## Verification
 
-ACTIONS: create (new skill), view (read full content), patch (update a section), edit (replace description + body), delete (remove skill).`;
+ACTIONS: create (new skill), view (read full content or list), patch (update a section by skill_id), edit (replace description + body by skill_id), delete (remove by skill_id).`;
 
 // ─── Interview prompt (onboarding) ───
 export const INTERVIEW_PROMPT = `You are conducting a brief onboarding interview with a new user. Your goal is to pre-fill their USER PROFILE so future sessions start with context instead of a blank slate.

package/src/handlers/auto-consolidate.ts

@@ -7,7 +7,7 @@
  * from disk after consolidation completes.
  */
 
-import type { ExtensionAPI } from "@mariozechner/pi-coding-agent";
+import type { ExtensionAPI } from "@earendil-works/pi-coding-agent";
 import { MemoryStore } from "../store/memory-store.js";
 import { CONSOLIDATION_PROMPT, ENTRY_DELIMITER } from "../constants.js";
 import type { ConsolidationResult } from "../types.js";

package/src/handlers/background-review.ts

@@ -7,7 +7,7 @@
  * keeping us within Pi's intended extension API.
  */
 
-import type { ExtensionAPI } from "@mariozechner/pi-coding-agent";
+import type { ExtensionAPI } from "@earendil-works/pi-coding-agent";
 import { MemoryStore } from "../store/memory-store.js";
 import { COMBINED_REVIEW_PROMPT } from "../constants.js";
 import type { MemoryConfig } from "../types.js";

package/src/handlers/correction-detector.ts

@@ -8,7 +8,7 @@
  * - Negative patterns: suppress even if a positive pattern matched
  */
 
-import type { ExtensionAPI } from "@mariozechner/pi-coding-agent";
+import type { ExtensionAPI } from "@earendil-works/pi-coding-agent";
 import { MemoryStore } from "../store/memory-store.js";
 import { DatabaseManager } from "../store/db.js";
 import {

package/src/handlers/index-sessions.ts

@@ -5,7 +5,7 @@
 import path from 'node:path';
 import fs from 'node:fs';
 import os from 'node:os';
-import type { ExtensionAPI, ExtensionCommandContext } from "@mariozechner/pi-coding-agent";
+import type { ExtensionAPI, ExtensionCommandContext } from "@earendil-works/pi-coding-agent";
 import { DatabaseManager } from '../store/db.js';
 import { indexAllSessions, getSessionStats } from '../store/session-indexer.js';
 

package/src/handlers/insights.ts

@@ -2,7 +2,7 @@
  * Insights command — /memory-insights shows what's stored in persistent memory.
  */
 
-import type { ExtensionAPI } from "@mariozechner/pi-coding-agent";
+import type { ExtensionAPI } from "@earendil-works/pi-coding-agent";
 import { MemoryStore } from "../store/memory-store.js";
 
 export function registerInsightsCommand(pi: ExtensionAPI, store: MemoryStore, projectStore: MemoryStore | null, projectName: string): void {

package/src/handlers/interview.ts

@@ -6,7 +6,7 @@
  * zero value until multiple sessions accumulate facts organically.
  */
 
-import type { ExtensionAPI } from "@mariozechner/pi-coding-agent";
+import type { ExtensionAPI } from "@earendil-works/pi-coding-agent";
 import { MemoryStore } from "../store/memory-store.js";
 import { INTERVIEW_PROMPT } from "../constants.js";
 

package/src/handlers/learn-memory.ts

@@ -2,7 +2,7 @@
  * Learn memory tool command — /learn-memory-tool teaches users about the memory system.
  */
 
-import type { ExtensionAPI, ExtensionCommandContext } from "@mariozechner/pi-coding-agent";
+import type { ExtensionAPI, ExtensionCommandContext } from "@earendil-works/pi-coding-agent";
 
 export function registerLearnMemoryCommand(pi: ExtensionAPI): void {
   pi.registerCommand("learn-memory-tool", {
@@ -34,7 +34,7 @@ export function registerLearnMemoryCommand(pi: ExtensionAPI): void {
         lines.push("  🧠 Memory       │ MEMORY.md     │ 5,000 chars");
         lines.push("  👤 User Profile │ USER.md       │ 5,000 chars");
         lines.push("  ⚠️  Failures     │ failures.md   │ 10,000 chars");
-        lines.push("  📚 Skills       │ skills/*.md   │ Unlimited");
+        lines.push("  📚 Skills       │ Pi-native skill dirs │ Unlimited");
         lines.push("  💾 Extended     │ sessions.db   │ Unlimited");
         lines.push("");
         lines.push("  Memory:   Facts — env details, project conventions, tool quirks");
@@ -126,7 +126,7 @@ export function registerLearnMemoryCommand(pi: ExtensionAPI): void {
         lines.push("  8. Session ends       → Final flush");
         lines.push("");
         lines.push("  Legacy mode: set memoryMode=\"legacy-inject\" to restore full");
-        lines.push("  MEMORY.md, USER.md, project memory, failure, and skill prompt blocks.");
+        lines.push("  MEMORY.md, USER.md, project memory, and failure prompt blocks.");
       }
 
       if (section.startsWith("🏗️")) {
@@ -153,7 +153,7 @@ export function registerLearnMemoryCommand(pi: ExtensionAPI): void {
         lines.push("  │ memory_search(\"auth\", cat:\"failure\")│");
         lines.push("  └─────────────────────────────────────┘");
         lines.push("");
-        lines.push("  Legacy mode can still inject full memory/skill blocks for users");
+        lines.push("  Legacy mode can still inject full memory blocks for users");
         lines.push("  who explicitly opt into memoryMode=\"legacy-inject\".");
       }
 

package/src/handlers/preview-context.ts

@@ -1,11 +1,10 @@
 /**
  * Preview context command — /memory-preview-context shows the policy-only prompt
- * or legacy memory/skill blocks appended to the system prompt.
+ * or legacy memory blocks appended to the system prompt.
  */
 
-import type { ExtensionAPI } from "@mariozechner/pi-coding-agent";
+import type { ExtensionAPI } from "@earendil-works/pi-coding-agent";
 import { MemoryStore } from "../store/memory-store.js";
-import { SkillStore } from "../store/skill-store.js";
 import { resolveMemoryPolicyPrompt } from "../prompt-context.js";
 import type { MemoryConfig } from "../types.js";
 
@@ -13,12 +12,11 @@ export function registerPreviewContextCommand(
   pi: ExtensionAPI,
   store: MemoryStore,
   projectStore: MemoryStore | null,
-  skillStore: SkillStore,
   projectName: string,
   config: Pick<MemoryConfig, "memoryMode" | "memoryPolicyStyle" | "memoryPolicyCustomText"> = { memoryMode: "policy-only" },
 ): void {
   pi.registerCommand("memory-preview-context", {
-    description: "Preview the memory policy or legacy memory/skill context blocks",
+    description: "Preview the memory policy or legacy memory context blocks",
     handler: async (_args, ctx) => {
       if (config.memoryMode === "policy-only") {
         const policyPrompt = resolveMemoryPolicyPrompt(config);
@@ -48,7 +46,6 @@ export function registerPreviewContextCommand(
 
       const memoryBlock = store.formatForSystemPrompt();
       const projectBlock = projectStore ? projectStore.formatProjectBlock(projectName) : "";
-      const skillIndex = await skillStore.formatIndexForSystemPrompt();
 
       const lines: string[] = [];
       lines.push("");
@@ -56,7 +53,7 @@ export function registerPreviewContextCommand(
       lines.push("  ║        👀 Injected Context Preview          ║");
       lines.push("  ╚══════════════════════════════════════════════╝");
       lines.push("");
-      lines.push("  This is the memory/skill context appended to the system prompt.");
+      lines.push("  This is the memory context appended to the system prompt.");
       lines.push("  (Core hidden system instructions are NOT shown.)");
       lines.push("");
 
@@ -76,16 +73,9 @@ export function registerPreviewContextCommand(
         lines.push("");
       }
 
-      if (skillIndex) {
-        blockCount++;
-        lines.push("  ── SKILL INDEX ─────────────────────────────────────────────");
-        lines.push(skillIndex);
-        lines.push("");
-      }
-
       if (blockCount === 0) {
         lines.push("  No memory context blocks are currently injected.");
-        lines.push("  Add memory entries or skills, then run this command again.");
+        lines.push("  Add memory entries, then run this command again.");
         lines.push("");
       }
 

package/src/handlers/session-flush.ts

@@ -4,7 +4,7 @@
  * See PLAN.md → "Hermes Source File Reference Map" for source lines.
  */
 
-import type { ExtensionAPI } from "@mariozechner/pi-coding-agent";
+import type { ExtensionAPI } from "@earendil-works/pi-coding-agent";
 import { MemoryStore } from "../store/memory-store.js";
 import { FLUSH_PROMPT } from "../constants.js";
 import type { MemoryConfig } from "../types.js";

package/src/handlers/skill-auto-trigger.ts

@@ -5,13 +5,22 @@
  * This implements Hermes' "self-evaluation checkpoint" pattern.
  */
 
-import type { ExtensionAPI } from "@mariozechner/pi-coding-agent";
+import * as path from "node:path";
+import { fileURLToPath } from "node:url";
+import type { ExtensionAPI } from "@earendil-works/pi-coding-agent";
 import { MemoryStore } from "../store/memory-store.js";
 import { SkillStore } from "../store/skill-store.js";
 import { COMBINED_REVIEW_PROMPT, DEFAULT_SKILL_TRIGGER_TOOL_CALLS, ENTRY_DELIMITER } from "../constants.js";
 import type { MemoryConfig } from "../types.js";
 import { getMessageText } from "../types.js";
 
+const proceduralSkillCreatorPath = path.join(
+  path.dirname(fileURLToPath(import.meta.url)),
+  "..",
+  "skills",
+  "procedural-skill-creator",
+);
+
 export function setupSkillAutoTrigger(
   pi: ExtensionAPI,
   store: MemoryStore,
@@ -71,10 +80,21 @@ export function setupSkillAutoTrigger(
 
       const currentMemory = store.getMemoryEntries().join(ENTRY_DELIMITER);
       const skillIndex = await skillStore.loadIndex();
-      const skillSummary = skillIndex.map((s) => `${s.fileName}: ${s.name} - ${s.description}`).join("\n");
+      const skillSummary = skillIndex
+        .map((skill) => `${skill.skillId} [${skill.scope}]: ${skill.displayName || skill.name} - ${skill.description}`)
+        .join("\n");
+
+      const activeProjectName = skillStore.getProjectName();
+      const scopeLine = activeProjectName
+        ? `Active project context: '${activeProjectName}'. If workflow details depend on this project, use scope='project'; otherwise use scope='global'.`
+        : "No active project context detected. Use scope='global' unless the workflow is clearly project-specific.";
 
       const prompt = [
         "This was a complex task that required multiple tool calls. Extract any reusable procedures as skills.",
+        "A bundled skill named procedural-skill-creator is loaded for you. Read and follow it before deciding whether to create or patch a skill.",
+        "Always pass scope explicitly when creating a skill: scope='global' or scope='project'.",
+        "Choose scope='global' for transferable procedures and scope='project' when the workflow depends on this repo's paths, scripts, architecture, deploy steps, or conventions.",
+        scopeLine,
         "",
         "--- Existing Skills ---",
         skillSummary || "(none)",
@@ -86,11 +106,11 @@ export function setupSkillAutoTrigger(
         recentParts.join("\n\n"),
         "",
         "If a skill should be created, use the skill tool with action 'create'.",
-        "If a related skill already exists, use 'patch' to update it.",
+        "If a related skill already exists, use 'patch' with its skill_id to update it.",
         "If nothing reusable happened, say 'Nothing to extract.' and stop.",
       ].join("\n");
 
-      const result = await pi.exec("pi", ["-p", "--no-session", prompt], {
+      const result = await pi.exec("pi", ["-p", "--no-session", "--skill", proceduralSkillCreatorPath, prompt], {
         signal: ctx.signal,
         timeout: 60000,
       });