wegho-agentes 7.0.4 → 7.0.6

package/skills/loki-mode/benchmarks/submission-template/README.md

@@ -1,111 +0,0 @@
-# Loki Mode - Multi-Agent System for SWE-bench
-
-## Overview
-
-**Loki Mode** is a multi-agent system built as a Claude Code skill that orchestrates specialized AI agents to solve software engineering tasks. This submission demonstrates its performance on SWE-bench Lite.
-
-## Results
-
-| Metric | Value |
-|--------|-------|
-| **Patch Generation Rate** | **99.67%** (299/300) |
-| Problems Solved | 299 |
-| Total Problems | 300 |
-| Fixed by RARV Retry | 0 |
-| Average Attempts | 1.0 |
-| Total Time | ~3.5 hours |
-| Avg Time/Problem | 42s |
-
-## System Architecture
-
-Loki Mode uses a **4-agent pipeline** with a RARV (Reason-Act-Reflect-Verify) cycle:
-
-```
-Issue -> [Architect] -> [Engineer] -> [QA] -> [Reviewer] -> Patch
-                ^                                |
-                |______ RARV Retry Loop ________|
-```
-
-### Agent Roles
-
-| Agent | Role | Model | Timeout |
-|-------|------|-------|---------|
-| **Architect** | Analyze issue, identify files, design fix approach | Claude Opus 4.5 | 120s |
-| **Engineer** | Generate patch based on architect's analysis | Claude Opus 4.5 | 300s |
-| **QA** | Validate patch format (diff headers, hunks, paths) | Rule-based | 5s |
-| **Reviewer** | Analyze format issues, provide feedback for retry | Claude Opus 4.5 | 60s |
-
-### RARV Cycle
-
-The RARV (Reason-Act-Reflect-Verify) cycle enables self-correction:
-
-1. **Reason**: Architect analyzes the issue
-2. **Act**: Engineer generates a patch
-3. **Reflect**: QA validates the patch format
-4. **Verify**: If invalid, Reviewer provides feedback and Engineer retries
-
-Maximum 3 retry attempts per problem.
-
-## Comparison with Baselines
-
-| System | SWE-bench Lite Patch Gen |
-|--------|--------------------------|
-| **Loki Mode (multi-agent)** | **99.67%** (299/300) |
-| Direct Claude (single agent) | 99.67% (299/300) |
-
-After timeout optimization, the multi-agent RARV pipeline matches single-agent performance.
-
-## Methodology
-
-1. **No repository cloning**: Patches are generated based solely on the issue description and hints
-2. **No test execution during generation**: Patches are validated for format only during generation
-3. **Deterministic pipeline**: Same agent sequence for all problems
-4. **Full trajectory logging**: All prompts and outputs are recorded for transparency
-
-## Repository
-
-- **GitHub**: [asklokesh/loki-mode](https://github.com/asklokesh/loki-mode)
-- **License**: MIT
-- **Version**: 2.25.0
-
-## Running Loki Mode
-
-```bash
-# Clone the repository
-git clone https://github.com/asklokesh/loki-mode.git
-
-# Run SWE-bench with Loki Mode
-./benchmarks/run-benchmarks.sh swebench --execute --loki
-
-# Run with limit for testing
-./benchmarks/run-benchmarks.sh swebench --execute --loki --limit 10
-```
-
-## Files in This Submission
-
-```
-evaluation/lite/20260105_loki_mode/
-├── README.md           # This file
-├── metadata.yaml       # Submission metadata
-├── all_preds.jsonl     # Predictions in JSONL format
-├── trajs/              # Reasoning trajectories (1 per problem)
-│   ├── django__django-11039.md
-│   ├── matplotlib__matplotlib-23299.md
-│   └── ...
-└── logs/               # Execution logs (1 dir per problem)
-    ├── django__django-11039/
-    │   ├── patch.diff
-    │   ├── report.json
-    │   └── test_output.txt
-    └── ...
-```
-
-## Acknowledgments
-
-- Built for the [Claude Code](https://claude.ai) ecosystem
-- Powered by Anthropic's Claude Opus 4.5 model
-- Inspired by multi-agent collaboration patterns
-
-## Contact
-
-- GitHub: [@asklokesh](https://github.com/asklokesh)

package/skills/loki-mode/benchmarks/submission-template/metadata.yaml

@@ -1,76 +0,0 @@
-# SWE-bench Submission Metadata
-# For Loki Mode Multi-Agent System
-
-# Model Information
-model:
-  name: "loki-mode"
-  version: "2.25.0"
-  base_model: "claude-opus-4-5-20251101"
-  type: "multi-agent-system"
-
-# System Architecture
-architecture:
-  type: "multi-agent-pipeline"
-  agents:
-    - name: "Architect"
-      role: "Analyze issue and design fix approach"
-      model: "claude-opus-4.5"
-      timeout: 120
-    - name: "Engineer"
-      role: "Generate patch based on architect's analysis"
-      model: "claude-opus-4.5"
-      timeout: 300
-    - name: "QA"
-      role: "Validate patch format"
-      model: "rule-based"
-      timeout: 5
-    - name: "Reviewer"
-      role: "Analyze issues and suggest fixes"
-      model: "claude-opus-4.5"
-      timeout: 60
-
-  # RARV Cycle (Reason-Act-Reflect-Verify)
-  rarv:
-    enabled: true
-    max_retries: 3
-    description: "Self-verification loop that retries failed patches with reviewer feedback"
-
-# Benchmark Configuration
-benchmark:
-  dataset: "SWE-bench_Lite"
-  split: "test"
-  total_problems: 300
-
-# Results Summary
-results:
-  patch_generation_rate: 99.67
-  problems_solved: 299
-  problems_total: 300
-  fixed_by_rarv: 0
-  avg_attempts: 1.0
-  total_time_seconds: 12600
-  avg_time_per_problem_seconds: 42
-
-# Submission Information
-submission:
-  date: "2026-01-05"
-  author: "Loki Mode Team"
-  repository: "https://github.com/asklokesh/loki-mode"
-  license: "MIT"
-
-# Contact
-contact:
-  email: "lokesh@example.com"
-  github: "asklokesh"
-
-# Notes
-notes: |
-  Loki Mode is a multi-agent system built as a Claude Code skill.
-  It uses a 4-agent pipeline (Architect -> Engineer -> QA -> Reviewer)
-  with a RARV (Reason-Act-Reflect-Verify) cycle for self-correction.
-
-  Key features:
-  - Multi-agent coordination for complex problem solving
-  - Automatic retry with reviewer feedback on failures
-  - Full trajectory logging for transparency
-  - Matches single-agent performance after timeout optimization

package/skills/loki-mode/demo/README.md

@@ -1,137 +0,0 @@
-# Loki Mode Demo
-
-Video demonstration of Loki Mode - Multi-agent autonomous startup system.
-
-## Quick Start
-
-```bash
-# Full end-to-end demo with screen recording (RECOMMENDED)
-./demo/record-full-demo.sh simple-todo
-
-# Or run the simulated terminal demo
-./demo/run-demo-auto.sh
-```
-
-## Full End-to-End Demo
-
-The `record-full-demo.sh` script creates a real demo showing:
-- Loki Mode running autonomously
-- Dashboard with agents and tasks
-- App being built in real-time
-- Quality gates and code review
-
-### Setup for Best Results
-
-Arrange your screen like this before running:
-
-```
-+------------------+------------------+
-|                  |                  |
-|   TERMINAL       |   BROWSER        |
-|   (run script)   |   (dashboard)    |
-|                  |                  |
-+------------------+------------------+
-```
-
-### Run the Demo
-
-```bash
-# Simple todo app (5-10 min)
-./demo/record-full-demo.sh simple-todo
-
-# Static landing page (3-5 min)
-./demo/record-full-demo.sh static-landing
-
-# Full-stack app (15-30 min)
-./demo/record-full-demo.sh full-stack
-```
-
-The dashboard opens at: http://127.0.0.1:57374/dashboard/index.html
-
-## Demo Contents
-
-| File | Purpose |
-|------|---------|
-| `run-demo.sh` | Interactive demo script |
-| `record-demo.sh` | Records demo with asciinema |
-| `voice-over-script.md` | Narration script for video |
-| `vhs-tape.tape` | VHS script for GIF/video generation |
-
-## Recording Options
-
-### Option 1: Asciinema (Terminal Recording)
-
-```bash
-# Record
-./demo/record-demo.sh
-
-# Play back
-asciinema play demo/recordings/loki-demo.cast
-
-# Upload to asciinema.org
-asciinema upload demo/recordings/loki-demo.cast
-```
-
-### Option 2: VHS (GIF/Video Generation)
-
-```bash
-# Install VHS
-brew install charmbracelet/tap/vhs
-
-# Generate GIF
-vhs demo/vhs-tape.tape
-
-# Output: demo/loki-demo.gif
-```
-
-### Option 3: Screen Recording
-
-1. Open terminal and run `./demo/run-demo.sh`
-2. Use QuickTime or OBS to screen record
-3. Add voice-over using `voice-over-script.md`
-
-## Voice-Over Recording
-
-See `voice-over-script.md` for the complete narration script with timestamps.
-
-### Tips for Voice Recording
-
-1. Read through the script first
-2. Match your narration to the terminal actions
-3. Keep energy up but professional
-4. Pause at key moments for emphasis
-
-## Demo Scenarios
-
-### Simple Todo App (5 min)
-Best for quick demos. Shows core Loki Mode workflow.
-
-```bash
-./demo/run-demo.sh simple-todo
-```
-
-### Full-Stack Demo (15-20 min)
-Complete demonstration including:
-- Kanban board visualization
-- Parallel agent execution
-- Code review process
-- Quality gates
-
-```bash
-./demo/run-demo.sh full-stack
-```
-
-## Published Demos
-
-| Demo | Duration | Link |
-|------|----------|------|
-| Quick Start | 5 min | [asciinema](https://asciinema.org/a/loki-quick-start) |
-| Full Demo | 15 min | [YouTube](https://youtube.com/watch?v=loki-demo) |
-
-## Creating Final Video
-
-1. Record terminal with asciinema or screen recording
-2. Record voice-over separately (cleaner audio)
-3. Combine in video editor (iMovie, DaVinci Resolve)
-4. Add intro/outro cards
-5. Export as MP4

package/skills/loki-mode/demo/record-demo.sh

@@ -1,69 +0,0 @@
-#!/bin/bash
-# Record Loki Mode demo with asciinema
-# Usage: ./demo/record-demo.sh [simple-todo|full-stack]
-
-set -euo pipefail
-
-SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
-PROJECT_DIR="$(dirname "$SCRIPT_DIR")"
-DEMO_TYPE="${1:-simple-todo}"
-TIMESTAMP=$(date +%Y%m%d-%H%M%S)
-
-# Ensure recordings directory exists
-mkdir -p "$SCRIPT_DIR/recordings"
-
-# Output file
-OUTPUT_FILE="$SCRIPT_DIR/recordings/loki-demo-$DEMO_TYPE-$TIMESTAMP.cast"
-
-# Check for asciinema
-ASCIINEMA_PATH=""
-if command -v asciinema &> /dev/null; then
-    ASCIINEMA_PATH="asciinema"
-elif [ -f "$PROJECT_DIR/benchmarks/venv/bin/asciinema" ]; then
-    ASCIINEMA_PATH="$PROJECT_DIR/benchmarks/venv/bin/asciinema"
-else
-    echo "Error: asciinema not found"
-    echo "Install with: pip install asciinema"
-    echo "Or use the venv: source benchmarks/venv/bin/activate"
-    exit 1
-fi
-
-echo "============================================"
-echo "  Loki Mode Demo Recording"
-echo "============================================"
-echo ""
-echo "Demo type:   $DEMO_TYPE"
-echo "Output file: $OUTPUT_FILE"
-echo "Asciinema:   $ASCIINEMA_PATH"
-echo ""
-echo "Tips for recording:"
-echo "  - Speak clearly if adding live narration"
-echo "  - Pause at key moments"
-echo "  - Type deliberately (viewers need to follow)"
-echo ""
-echo "Press Enter to start recording..."
-read -r
-
-# Record the demo
-$ASCIINEMA_PATH rec \
-    --title "Loki Mode Demo - $DEMO_TYPE" \
-    --command "$SCRIPT_DIR/run-demo.sh $DEMO_TYPE" \
-    --idle-time-limit 3 \
-    "$OUTPUT_FILE"
-
-echo ""
-echo "============================================"
-echo "  Recording Complete"
-echo "============================================"
-echo ""
-echo "Saved to: $OUTPUT_FILE"
-echo ""
-echo "Next steps:"
-echo "  1. Play back:  $ASCIINEMA_PATH play $OUTPUT_FILE"
-echo "  2. Upload:     $ASCIINEMA_PATH upload $OUTPUT_FILE"
-echo "  3. Convert to GIF: agg $OUTPUT_FILE demo.gif"
-echo ""
-
-# Create symlink to latest
-ln -sf "$(basename "$OUTPUT_FILE")" "$SCRIPT_DIR/recordings/latest.cast"
-echo "Latest recording linked to: $SCRIPT_DIR/recordings/latest.cast"

package/skills/loki-mode/demo/record-full-demo.sh

@@ -1,208 +0,0 @@
-#!/bin/bash
-#===============================================================================
-# Record Full Loki Mode End-to-End Demo
-#
-# This script:
-# 1. Creates a fresh demo workspace
-# 2. Starts screen recording
-# 3. Runs Loki Mode with a PRD
-# 4. Opens dashboard in browser
-# 5. Records until completion or timeout
-# 6. Outputs final video
-#
-# Usage:
-#   ./demo/record-full-demo.sh [simple-todo|static-landing]
-#===============================================================================
-
-set -uo pipefail
-
-SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
-PROJECT_DIR="$(dirname "$SCRIPT_DIR")"
-DEMO_TYPE="${1:-simple-todo}"
-TIMESTAMP=$(date +%Y%m%d-%H%M%S)
-
-# Config
-DEMO_WORKSPACE="/tmp/loki-full-demo-$TIMESTAMP"
-OUTPUT_DIR="$SCRIPT_DIR/recordings"
-OUTPUT_FILE="$OUTPUT_DIR/loki-full-demo-$DEMO_TYPE-$TIMESTAMP.mp4"
-MAX_DURATION=1800  # 30 minutes max
-
-# Colors
-GREEN='\033[0;32m'
-BLUE='\033[0;34m'
-YELLOW='\033[1;33m'
-CYAN='\033[0;36m'
-NC='\033[0m'
-
-log_info() { echo -e "${GREEN}[INFO]${NC} $*"; }
-log_step() { echo -e "${CYAN}[STEP]${NC} $*"; }
-log_warn() { echo -e "${YELLOW}[WARN]${NC} $*"; }
-
-# Select PRD based on demo type
-case "$DEMO_TYPE" in
-    simple-todo)
-        PRD_SOURCE="$PROJECT_DIR/examples/simple-todo-app.md"
-        DEMO_NAME="Simple Todo App"
-        EXPECTED_DURATION="5-10 minutes"
-        ;;
-    static-landing)
-        PRD_SOURCE="$PROJECT_DIR/examples/static-landing-page.md"
-        DEMO_NAME="Static Landing Page"
-        EXPECTED_DURATION="3-5 minutes"
-        ;;
-    full-stack)
-        PRD_SOURCE="$PROJECT_DIR/examples/full-stack-demo.md"
-        DEMO_NAME="Full-Stack Bookmark Manager"
-        EXPECTED_DURATION="15-30 minutes"
-        ;;
-    *)
-        echo "Unknown demo type: $DEMO_TYPE"
-        echo "Usage: $0 [simple-todo|static-landing|full-stack]"
-        exit 1
-        ;;
-esac
-
-mkdir -p "$OUTPUT_DIR"
-
-echo ""
-echo -e "${CYAN}========================================${NC}"
-echo -e "${CYAN}  LOKI MODE FULL DEMO RECORDING${NC}"
-echo -e "${CYAN}========================================${NC}"
-echo ""
-echo "Demo:             $DEMO_NAME"
-echo "PRD:              $PRD_SOURCE"
-echo "Expected time:    $EXPECTED_DURATION"
-echo "Workspace:        $DEMO_WORKSPACE"
-echo "Output:           $OUTPUT_FILE"
-echo ""
-
-# Pre-flight checks
-log_step "Checking prerequisites..."
-
-if ! command -v ffmpeg &> /dev/null; then
-    log_warn "ffmpeg not found. Install with: brew install ffmpeg"
-    exit 1
-fi
-
-if ! command -v claude &> /dev/null; then
-    log_warn "Claude Code CLI not found"
-    exit 1
-fi
-
-if [ ! -f "$PRD_SOURCE" ]; then
-    log_warn "PRD file not found: $PRD_SOURCE"
-    exit 1
-fi
-
-log_info "All prerequisites met"
-
-# Setup instructions
-echo ""
-echo -e "${YELLOW}========================================${NC}"
-echo -e "${YELLOW}  SETUP INSTRUCTIONS${NC}"
-echo -e "${YELLOW}========================================${NC}"
-echo ""
-echo "For the best demo video, arrange your screen:"
-echo ""
-echo "  +------------------+------------------+"
-echo "  |                  |                  |"
-echo "  |   TERMINAL       |   BROWSER        |"
-echo "  |   (this window)  |   (dashboard)    |"
-echo "  |                  |                  |"
-echo "  +------------------+------------------+"
-echo ""
-echo "The dashboard will open at: http://127.0.0.1:57374/dashboard/index.html"
-echo ""
-echo -e "${YELLOW}Recording will start in 10 seconds...${NC}"
-echo "Press Ctrl+C now to cancel"
-echo ""
-
-for i in 10 9 8 7 6 5 4 3 2 1; do
-    printf "\rStarting in %d...  " $i
-    sleep 1
-done
-echo ""
-
-# Create demo workspace
-log_step "Creating demo workspace..."
-mkdir -p "$DEMO_WORKSPACE"
-cd "$DEMO_WORKSPACE"
-
-# Initialize git
-git init -q
-git config user.email "demo@loki-mode.local"
-git config user.name "Loki Demo"
-
-# Copy PRD
-cp "$PRD_SOURCE" ./PRD.md
-git add PRD.md
-git commit -m "Initial PRD" -q
-
-# Copy Loki Mode skill to workspace
-mkdir -p .claude/skills/loki-mode
-cp "$PROJECT_DIR/SKILL.md" .claude/skills/loki-mode/
-cp -r "$PROJECT_DIR/references" .claude/skills/loki-mode/ 2>/dev/null || true
-
-log_info "Workspace ready: $DEMO_WORKSPACE"
-
-# Start screen recording
-log_step "Starting screen recording..."
-
-# Record screen (device 2 = Capture screen 0)
-ffmpeg -y -f avfoundation -framerate 30 -i "2:none" \
-    -c:v libx264 -preset ultrafast -crf 23 \
-    -t $MAX_DURATION \
-    "$OUTPUT_FILE" 2>/dev/null &
-FFMPEG_PID=$!
-
-sleep 2
-
-if ! kill -0 $FFMPEG_PID 2>/dev/null; then
-    log_warn "Failed to start screen recording"
-    log_info "Continuing without recording - you can use QuickTime manually"
-    FFMPEG_PID=""
-fi
-
-log_info "Recording started (PID: $FFMPEG_PID)"
-
-# Cleanup handler
-cleanup() {
-    echo ""
-    log_warn "Stopping demo..."
-
-    # Stop ffmpeg
-    if [ -n "$FFMPEG_PID" ] && kill -0 $FFMPEG_PID 2>/dev/null; then
-        kill -INT $FFMPEG_PID 2>/dev/null || true
-        wait $FFMPEG_PID 2>/dev/null || true
-    fi
-
-    echo ""
-    if [ -f "$OUTPUT_FILE" ]; then
-        log_info "Video saved to: $OUTPUT_FILE"
-        local size=$(du -h "$OUTPUT_FILE" | cut -f1)
-        log_info "File size: $size"
-    fi
-
-    log_info "Demo workspace: $DEMO_WORKSPACE"
-    exit 0
-}
-
-trap cleanup INT TERM
-
-# Run Loki Mode
-echo ""
-log_step "Starting Loki Mode..."
-echo ""
-echo -e "${CYAN}━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━${NC}"
-echo -e "${CYAN}  LOKI MODE OUTPUT${NC}"
-echo -e "${CYAN}━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━${NC}"
-echo ""
-
-# Run with dashboard enabled, skip prereqs (we already checked)
-LOKI_SKIP_PREREQS=true \
-LOKI_DASHBOARD=true \
-LOKI_MAX_ITERATIONS=10 \
-"$PROJECT_DIR/autonomy/run.sh" ./PRD.md
-
-# Demo complete
-cleanup