@agent-smith/feat-agents 0.0.1

package/dist/skills/update-project-nav/SKILL.md

@@ -0,0 +1,97 @@
+---
+name: update-project-nav
+description: use when asked to update the project navigation map documentation
+---
+
+This skill describes the workflow to update the project navigation map file (`.agents/documentation/project-nav.md`). This is the canonical reference for AI agents to understand and navigate the entire project.
+
+## Workflow
+
+1. **Read existing file** — Load `.agents/documentation/project-nav.md` from the project root. Note the current sections and content.
+
+2. **Detect changes** — Walk the directory tree, identify what has changed since the last update:
+   - New/removed repos or packages
+   - New/modified plugins
+   - UI component changes
+   - New documentation files
+   - Architecture changes (new patterns, dependencies)
+
+3. **Update sections** — Go through each section and update as needed. Preserve the existing format and style.
+
+### Section Update Rules
+
+#### 1. Project Overview
+- Update the repos table if repos are added/removed/renamed
+- Update the purpose descriptions if functionality changes
+- Keep the 4-column table format: `| Repo | Path | Purpose |`
+
+#### 2. Architecture Principles
+- Add new principles when architectural patterns emerge
+- Remove obsolete principles
+- Update "Key Files" column if file locations change
+- Maintain the 3-column table: `| Principle | Detail | Key Files |`
+
+#### 3. Dependency Graph
+- Regenerate ASCII art if dependency relationships change
+- Update the prose explanation below the graph
+- Keep the visual format consistent (boxes, arrows, labels)
+
+#### 4. Runtime Packages
+- Add new packages with: Purpose, Key files, Key types/classes
+- Update existing packages if API changes
+- Remove deprecated packages
+- Format: `### @agent-smith/<name> — <short description>` followed by bullet points
+
+#### 5. Server
+- Update route table with new/changed routes
+- Add new handler files
+- Update patterns if confirmation/async flow changes
+
+#### 6. Plugins
+- Add new plugins to the table
+- Update plugin categories if reorganized
+- Keep format: `| Plugin | Category | Purpose | Key File(s) |`
+
+#### 7. UI
+- Update component/service tables
+- Add new themes, apps, or extensions
+- Note any architectural changes (Vue version, PrimeVue updates)
+
+#### 8. Apps
+- Document new apps added to the dashboard
+- Update existing app descriptions if functionality changes
+
+#### 9. Code Snippets
+- Add new patterns for new APIs or features
+- Update existing snippets if API signatures change
+- Keep language-appropriate examples (YAML, JS, TS)
+
+#### 10. Navigation Quick Reference
+- Add new task→path mappings
+- Update paths if files moved
+- Keep the `| Task | Go To |` table format
+
+#### 11. Documentation Links
+- Add new documentation resources
+- Remove links to deleted docs
+- Verify all paths still exist
+
+#### 12. Key Conventions & Patterns
+- Add new conventions discovered in the codebase
+- Update existing conventions if they change
+- Keep the `| Convention | Detail |` table format
+
+4. **Cross-reference check** — Ensure no content duplication:
+   - Project description, architecture patterns, quick references, and code snippets live ONLY in `project-nav.md`
+   - Per-module technical details (entry points, key files, dependencies) live in `codebase-summary.md`
+   - No redundancy between files
+
+5. **Write the updated file** — Preserve the header note: `> **Purpose**: Single-reference map for AI coding agents...`
+
+## Rules
+
+- **Information-dense**: Use tables, bullets, one-line descriptions
+- **No redundancy**: Each piece of information lives in exactly one file
+- **Preserve format**: Do not change section structure or table formats
+- **Canonical source**: `project-nav.md` is the single source of truth for project overview, architecture, and navigation
+- **Language-agnostic**: Adapt to the project's language and ecosystem

package/dist/skills/use-tasks/SKILL.md

@@ -0,0 +1,103 @@
+---
+name: use-tasks
+description: reference guide for the task skills ecosystem — explains how tasks work and which skill to use
+---
+
+# Task Skills Reference Guide
+
+This skill provides an overview of the task system. Use it to understand which skill to load for your specific need.
+
+## Task Skills Ecosystem
+
+### Creation Skills
+| Skill | Purpose | When to Use |
+|-------|---------|-------------|
+| `create-task` | **Entry point** — Analyzes request, chooses mode, delegates | User asks to create a new task |
+| `task-mode` | **Mode selector** — Checklist to decide solo vs team | Need to determine execution mode |
+| `create-task-solo` | Creates task for single-agent execution | Task needs unified context (reading/analyzing data) |
+| `create-task-team` | Creates task for multi-agent execution | Task has independent phases needing different expertise |
+
+### Execution Skills
+| Skill | Purpose | When to Use |
+|-------|---------|-------------|
+| `execute-task-solo` | Execute a solo task (single agent runs all phases) | Running a task created with `create-task-solo` |
+| `execute-task-team` | Execute a team task (coordinator delegates to executors) | Coordinator running a task created with `create-task-team` |
+| `execute-task-phase` | Behavioral contract for executor agents | Executor agent running an assigned phase |
+
+## Workflows
+
+**Solo Task Workflow:**
+```
+create-task → task-mode → create-task-solo → execute-task-solo
+```
+
+**Team Task Workflow:**
+```
+create-task → task-mode → create-task-team → execute-task-team → execute-task-phase (executors)
+```
+
+## Task Location
+
+Tasks are stored in:
+- `.agents/tasks/[task-name]/` — Main project tasks
+- `[project-name].agents/tasks/[task-name]/` — Project-specific tasks
+
+## Task Directory Structure
+
+### Solo Task
+```
+[task-name]/
+├── goals.md              # Goals and success criteria
+├── plan.md               # Full execution plan with all phases
+├── state.md              # Progress tracking
+└── notes.md              # Optional: additional context/notes
+```
+
+### Team Task
+```
+[task-name]/
+├── goals.md                      # Goals and success criteria
+├── plan.md                       # Full execution plan (reference)
+├── state.md                      # Progress tracking
+├── coordinator-instructions.md   # Orchestration instructions
+└── phases/
+    ├── phase-1.md                # Individual phase files
+    └── phase-2.md
+```
+
+## Quick Reference: Which Skill to Load?
+
+| Your Goal | Load This Skill |
+|-----------|-----------------|
+| Create a new task | `create-task` |
+| Decide solo vs team mode | `task-mode` |
+| Execute a solo task | `execute-task-solo` |
+| Execute a team task (as coordinator) | `execute-task-team` |
+| Execute an assigned phase (as executor) | `execute-task-phase` |
+| Understand task structure | This skill (`use-tasks`) |
+
+## Task Lifecycle
+
+### 1. Create a Task
+- Check if `.agents/tasks/[task-name]/` exists; create if not
+- Task files should contain: name/description, goals, phase-by-phase plan with execution steps and success criteria, status tracking
+- Report to user before executing
+
+### 2. Execute a Task
+- Read `state.md` to check progress
+- Execute phases one by one (never skip ahead)
+- Verify success criteria after each step
+- Update `state.md` after each completed phase
+
+### 3. After Completion
+- Ask user if everything is satisfactory
+- Only delete task directory if user confirms
+- If incomplete: update `state.md` with what was done and what remains
+
+## Important Rules
+
+- Always execute steps one by one — never skip ahead
+- Verify success criteria after each step before proceeding
+- Update `state.md` after each completed step
+- Do not re-verify previously completed steps (check `state.md` first)
+- Ask user for confirmation before deleting task directory

package/dist/workflows/agent-smith-config-info.yml

@@ -0,0 +1,9 @@
+tool:
+  description: use this to only retrieve raw data in the Agent Smith config features database for agent, action, skill, command, adaptater, workflow config
+  arguments:
+    prompt:
+        description: the question to ask about the data
+steps:
+  - adaptater: agent-smith-db-getschema
+  - agent: agent-smith-sql
+  - action: db-execute-read-query

package/dist/workflows/agent-smith-db.yml

@@ -0,0 +1,9 @@
+tool:
+  description: and agent to retrieve configuration data in the Agent Smith config database
+  arguments:
+    prompt:
+        description: the question to ask about the data
+steps:
+  - adaptater: agent-smith-db-getschema
+  - agent: agent-smith-sql
+  - action: db-execute-read-query

package/dist/workflows/q.yml

@@ -0,0 +1,3 @@
+steps:
+  - adaptater: prequery
+  - agent: infer

package/dist/workflows/vision.yml

@@ -0,0 +1,3 @@
+steps:
+  - action: imgs2base64
+  - task: infer

package/package.json

@@ -0,0 +1,15 @@
+{
+  "name": "@agent-smith/feat-agents",
+  "version": "0.0.1",
+  "description": "AI agents for Agent Smith",
+  "repository": "https://github.com/lynxai-team/agent-smith-plugins",
+  "type": "module",
+  "files": [
+    "dist"
+  ],
+  "publishConfig": {
+    "access": "public",
+    "registry": "https://registry.npmjs.org/"
+  },
+  "license": "MIT"
+}