🎉 THE PRODUCTION RELEASE!

After months of intensive development and 90 validation runs across the WoNQ Matrix testing framework, QonQrete v1.0.0-stable is ready for prime time!

🔥 Critical Fix #1: Bulletproof Language Detection

The Problem (v0.9.x)

When using OpenAI as the AI provider, code blocks like ```py would create files named “py” instead of being skipped:

# AI output:                  # Result (WRONG):
```py                         → File "py" created
print("hello")                  (should have been skipped!)

The Fix (v1.0.0-stable)

The language_keywords set expanded from 23 entries to 400+ entries!

Provider Compatibility

🔥 Critical Fix #2: Enforced Briq Sensitivity

The Problem (v0.9.x)

Previously, briq_sensitivity was just a “hint” to the AI, resulting in wildly inconsistent outputs.

The Fix (v1.0.0-stable)

Briq counts are now ENFORCED with hard min/max ranges:

Enforcement Logic:

1. AI generates briqs with MANDATORY count prompt
2. System checks if count is within range
3. Too few → Retry with stronger prompt (up to 2 retries)
4. Too many → Merge consecutive briqs to meet max
5. Log compliance: [OK] Briq count 4 is within range [3-5]

🆕 New CLI Features

# Default: 4 cycles
./qonqrete.sh run

# Override to 6 cycles for complex projects
./qonqrete.sh run -c 6

# Combine with sensitivity for full control
./qonqrete.sh run -b 5 -c 6  # Complex project config
./qonqrete.sh run -b 7 -c 4  # Simple project config (default)

⚙️ Configuration Changes

🧪 WoNQ Matrix Validation

90 validation runs across every combination of sensitivity levels (0-9) and cycle counts (1-9):

╔═══════════════════════════════════════════════════════════════╗
║                 WoNQ MATRIX RESULTS                           ║
╠═══════════════════════════════════════════════════════════════╣
║ Total Runs:           90 (100% coverage)                      ║
║ Clean Completions:    90 (100% success)                       ║
║ Champion Score:       658 (sensitivity=3, cycle=7)            ║
║ Global Average:       554                                     ║
║ Scores ≥600:          35.6%                                   ║
╚═══════════════════════════════════════════════════════════════╝

Key Findings

1. Sweet Spot: Sensitivity 2-4 with 5-7 cycles produces the best results
2. Champion Config: sens=3, cycle=7 consistently hits peak performance
3. Avoid: Sensitivity ≥7 shows diminishing returns; Cycle 8 is a “death valley”

💰 Cost Efficiency

📊 Diff Summary

Total: +536 lines of production hardening

📋 Migration Guide

From v0.9.9-beta

1. No breaking changes - Drop-in replacement
2. Update config.yaml (optional):

   briq_sensitivity: 7  # was 5
   auto_cycle_limit: 4  # was 0
   

3. No container rebuild required

✅ Checklist

• All tests passing
• 90 WoNQ Matrix validation runs complete
• Multi-provider compatibility verified
• Documentation updated
• RELEASE-NOTES.md updated
• VERSION bumped to 1.0.0-stable
• No breaking changes from v0.9.9-beta

QonQrete is ready for prime time! 🗝️

🎯 Summary

Final stabilization release before v1.0.0. Fixes critical bugs discovered during multi-cycle autonomous builds, cleans up console output, and removes broken dependencies.

🐛 Bug Fixes

v0.9.8 (included)

v0.9.9

📝 Files Changed

Core Pipeline

• worqer/construqtor.py - Skeleton detection before write
• worqer/qontextor.py - Removed broken pycg, uses jedi only
• worqer/inspeqtor.py - Internal LoQal verification (Stage 3)
• worqspace/pipeline_config.yaml - Removed standalone loqal_verifier

Console Output

• qrane/qrane.py - Updated VISIBLE_KEYWORDS, BLOCKED_KEYWORDS, CONTENT_FILTER_PATTERNS

Dependencies

• requirements.txt - Removed pycg (broken package)

Documentation

• All docs updated to v0.9.9-beta
• doc/RELEASE-NOTES.md - Added v0.9.8 and v0.9.9 sections

✅ Verified Features

🔄 Migration

# Rebuild container (requirements.txt changed)
./qonqrete.sh init

🧪 Testing Done

• Multi-cycle autonomous builds (3+ cycles)
• Syntax validation on all Python files
• Verified skeleton detection works
• Verified console output is cleaner

📊 Console Output Before/After

Before v0.9.9:

〘aQQ〙『tasQleveler』⸎ ## 📋 Overview
〘aQQ〙『tasQleveler』⸎ ## 📦 Dependency Graph
〘aQQ〙『tasQleveler』⸎ # 🎯 Golden Path Test:
〘aQQ〙『inspeQtor』  ⸎ |-----|---------|-----|
〘aQQ〙『inspeQtor』  ⸎ -- Batch 1/1: 6 briqs --
〘aQQ〙『Qontextor』  ⸎ - [WARN] pycg not found...

After v0.9.9:

〘aQQ〙『tasQleveler』⸎ [TasqLeveler] ✅ Enhanced tasq: 180 lines
〘aQQ〙『inspeQtor』  ⸎ === InspeQtor v0.9.0 Complete: [SUCCESS] ===

Ready for v1.0.0 🚀

🚀 TasqLeveler Agent - Automatic Tasq Enhancement

The Big Idea: A well-structured tasq.md = dramatically better output quality. So why not let AI enhance your tasq automatically?

TasqLeveler is a new agent that runs ONCE on Cycle 1, BEFORE InstruQtor. It supercharges your tasq.md with:

🔧 Universal File Rule (s00permode)

The Problem: Previous “refinement mode” approach was too restrictive - it artificially limited what cycles 2+ could do, potentially blocking legitimate new file creation.

The Solution: Removed all mode-based logic. Replaced with ONE simple universal rule that applies to ALL cycles:

📁 File EXISTS in qodeyard? → MODIFY it or EXTEND it. Never recreate it. 📄 File DOESN’T EXIST yet? → CREATE it.

This prevents the rebuild-from-scratch bug while maintaining full creative freedom.

✅ Confirmed: Batched Reviews Working Perfectly!

Deep analysis of v0.8.7 production run (28 briqs, 7 cyQles) confirmed:

• Zero [UNKNOWN] assessments - batch parsing working correctly
• All [FAILURE] results are real AI assessments, not parse failures
• Retry mechanism working - briq027 failed once (Gemini API error), succeeded on attempt 2
• Cost efficiency achieved - 28 briqs reviewed in 3 batches for ~$0.01 total

🔧 LoQal Verifier Display Fix

Problem: “Running local validation…” was showing under construQtor in the event log, but it’s conceptually part of inspeQtor’s LoQal Verification stage.

Fix: Changed display to use [LoQal] prefix for clearer attribution.

🎯 Default Config Updates

Updated config.yaml with production-ready defaults, including upgrading construQtor to gemini-2.5-pro.

🐛 Bug Fix: Display Filter Was Completely Broken

Problem: ConstruQtor and InspeQtor status messages weren’t showing in the event log at all.

Root Cause: The is_content_line() function was being called BEFORE checking VISIBLE_KEYWORDS, and it was matching patterns in legitimate status lines.

Fix: Complete rewrite of display filter logic to prioritize VISIBLE_KEYWORDS and apply content filtering only to lines without visible keywords.

🚀 Major: Batched Reviews (90% Fewer API Calls!)

The Problem: v0.8.5 with per-briq reviews made 77+ API calls per cycle, burning $16+ with GPT-4.1.

The Solution: Batched reviews group multiple briqs into single API calls, resulting in up to 90% savings.

💰 New Default: Gemini 2.5 Flash-Lite

All agents now default to gemini-2.5-flash-lite ($0.10/$0.40 per 1M tokens), providing a 75-95% cost reduction compared to previous models.

🚨 TOKEN BURN FIX - CRITICAL

Problem Identified: Running InspeQtor with gpt-4.1 (not mini) on 77 briqs × 2 stages = ~154 AI calls burned ~$25 in a single run.

Root Cause: InspeQtor was configured with gpt-4.1 ($2.00/$8.00 per 1M tokens) instead of gpt-4.1-mini ($0.40/$1.60 per 1M tokens).

✨ New Features

1. Cost Estimation Display

InstruQtor and InspeQtor now show estimated costs before AI calls.

2. Cleaner Display Filter System

Completely overhauled the display filter to suppress noise from per-briq assessments and logs.

🐛 Bug Fixes

1. Fixed Empty/Invalid __init__.py Files Being Written

Problem: AI sometimes outputs empty code blocks resulting in files containing just ``` (markdown fence) instead of valid Python.

Fix: Improved code block regex parser to skip files with empty or invalid content.

2. Improved Import Resolution in LoQal Verifier

Problem: Import checker was flagging src.utils.logger as missing even when src/utils/logger.py existed.

Fix: Enhanced import resolution to search recursively for module names.

✨ New Features & Enhancements

1. Interleaved Pipeline (Build → Validate → Build → Validate)

ConstruQtor now processes each briq with interleaved validation, running local syntax and import checks after each build step.

2. Per-Briq ExeQ Summaries

ConstruQtor now writes execution summaries to exeq.d/cyqle{N}/ for each briq.

3. Smarter LoQal Verifier Skeleton Matching

Fixed false positive warnings for standard library imports and typing constructs.

✨ LoQal Verifier Integration

Integrated a deterministic local verification agent that runs after ConstruQtor.

What It Checks (NO AI Required)

1. Syntax Validation: python -m py_compile on all .py files
2. Import Resolution: Verifies local imports resolve to actual files
3. Skeleton Signature Matching: Ensures functions exist that are called elsewhere

Self-Healing Feedback Loop

Verification errors are appended to the reqap.md, which is read by InstruQtor in the next cycle, creating a self-healing feedback loop.

✨ ConstruQtor Per-Briq Retry

Added configurable retry mechanism for individual briq failures:

# config.yaml
agents:
  construqtor:
    max_attempts: 3           # Retry failed briqs up to 3 times
    stop_on_briq_fail: true   # Stop cycle if briq fails after all retries

• ConstruQtor now tracks attempts per briq
• Failed briqs are retried with error context appended to prompt
• Status output shows attempt count: [✅ SUCCESS] (attempts: 2)
• Configurable stop_on_briq_fail - set false to continue despite failures

🌀 Qontrabender - The Cache Bender

A new agent that manages hybrid caching with intelligent content classification:

• Variable Fidelity: Mixes MEAT (full code) + BONES (skeletons) based on file importance
• Policy-Driven Configuration: All behavior controlled via caching_policy.yaml
• Multiple Operational Modes: 6 pre-configured modes for different use cases
• Schema Validation: YAML validation prevents bad configuration from breaking the flow
• Improved Volatile Detection: Cycle-based, diff-based, git diff, and mtime fallback

📊 Triple-Core Memory System

QonQrete now features a Triple-Core Memory System:

This release introduces a major upgrade to the qontextor agent, enabling a fully local, deterministic, and highly detailed analysis of the codebase. This new “Local Qontextor Stack” significantly reduces reliance on AI for context generation, leading to massive cost savings, increased speed, and enhanced privacy.

✨ New Features & Major Enhancements

1. Fully Local qontextor Agent

The qontextor agent can now run in a completely local mode (provider: local in config.yaml), which is the new default. This mode uses a sophisticated stack of local analysis tools to build a deep understanding of the codebase without any AI calls.

2. The Local Qontextor Stack

The new local mode is powered by a multi-layered analysis stack:

• Python AST: For extracting the fundamental structure of the code (classes, functions, signatures).
• Docstrings & Verb Heuristics: To understand the purpose of code, either from existing documentation or by inferring it from function names.
• Jedi: For static analysis, providing type inference and cross-file relationship understanding.
• PyCG: To generate a comprehensive call graph, mapping out dependencies and execution flow.

3. Fast vs. Complex Local Modes

The local qontextor can be fine-tuned for speed or detail:

• local_mode: 'fast': Provides a very fast analysis using AST, Jedi, and heuristics.
• local_mode: 'complex': Enhances the analysis by using a local sentence-transformers model to create deep semantic embeddings of the code’s purpose.

4. qontextor CLI Helpers

The qontextor agent can now be invoked directly from the command line:

• python3 worqer/qontextor.py --query "<search_term>": Performs a semantic search for symbols.
• python3 worqer/qontextor.py --verb "<verb_pattern>": Finds symbols matching a verb pattern.
• python3 worqer/qontextor.py --ripple "<symbol_name>": Analyzes the ripple effect of changing a symbol.

5. Enforced Verb Usage for construQtor

The construQtor agent’s prompt has been updated to enforce strict naming conventions for functions and methods.

6. Increased Verbosity for Agents

qompressor and qontextor now provide more verbose output, printing each file they process.

🐛 Bug Fixes

• Fixed a NameError in the inspeqtor agent that was causing it to crash during the review phase.
• Fixed a NameError in the qontextor agent related to the extract_first_sentence function.
• Added a docker system prune command to qonqrete.sh to prevent “No space left on device” errors.

🚀 Performance & Cost

• Indexing Cost: Reduced to zero when using the local qontextor.
• Cost per Run: Up to 25x cheaper due to massive reduction in tokens sent to AI providers.
• Speed: Approximately 3x faster on average due to smaller prompts and local processing.

Added

• Dynamic Local Agent Loader: Implemented a dynamic local agent loader in qrane/qrane.py, allowing agents configured with provider: local in config.yaml to dynamically load and execute corresponding Python scripts from the worqer directory based on their model name.

Changed

• qrane.py: Modified run_orchestration to dynamically determine agent script paths for local providers.
• Dockerfile: Added npm install -g @qwen-code/qwen-code@latest to install the Qwen CLI tool.
• lib_ai.py: Modified _run_qwen to pass prompts to the qwen CLI via standard input instead of command-line arguments.

Fixed

• QWEN_API_KEY Environment Variable: Ensured qonqrete.sh passes QWEN_API_KEY to the container.
• construQtor Briq Processing: Improved construQtor’s handling of briqs.

v0.6.2-beta - 2025-12-18

Added

• “local” Provider: Implemented a “local” provider for offline agents like calqulator and qompressor.
• Qwen Model Testing: Tested qwen-turbo, qwen-coder, and qwen-max models.

Changed

• Default Briq Format: The instruqtor now defaults to a more reliable markdown-based format for briqs.

Fixed

• AI Reliability: The new markdown format for briqs significantly improves reliability.

v0.6.1-beta - 2025-12-16

Added

• Qwen Provider Integration: Integrated the Qwen AI provider into the system.
• New Documentation: Added extensive documentation on core concepts:
  • CONTEXT.md, MEMORY.md, MINDSTACK.md, MINDSTACK_ARCH.md, QWEN_90K_FIX.md, SKELETON.md

Changed

• Default Task: Updated worqspace/tasq.md to a more complex task for better testing.

This release introduces the Qompressor and Qontextor agents, forming a “Dual-Core” memory system that dramatically reduces cost and increases speed.

The Numbers

The Scenario: A medium-sized project (50 files, ~10,000 lines of code).

• Raw Size: ~100,000 Tokens.

Summary: You are paying 4% of the cost for 100% of the intelligence.

New Agents

• Qompressor (The Skeletonizer): Creates a low-token “skeleton” of the codebase in bloq.d. Provides architectural context with zero token cost.
• Qontextor (The Symbol Mapper): Uses AI to analyze skeletonized code and generate a detailed YAML map of symbols, purposes, and dependencies in qontext.d.
• CalQulator (The Cost Estimator): Analyzes briQ files to provide token and cost estimates for the upcoming construqtor cycle.
• FunQtions Library: New shared library qrane/lib_funqtions.py for common utility functions.

Bug Fixes

• Loader Spinner: Fixed blocking spinner animation using select.select() loop.
• Calqulator Output: Restored visibility of cost calculation table in headless mode.
• Briq Filenames: Corrected double underscores in briq filenames.

Added

• Comprehensive Test Suite: Introduced a new TESTS.md file outlining a full suite of functional tests, including a provider and model matrix, mode and briq sensitivity matrix, and edge/regression scenarios.
• Extensive Model Support: Added and verified support for a wide range of new models from OpenAI, Google, Anthropic, and DeepSeek.

Fixed

• Gemini API response.text error: Critical bug fix for crashes when Gemini API returned blocked/empty responses.
• Assessment Parsing: More robust parsing for the inspeqtor’s assessment output.

v0.4.9-alpha - 2025-12-07

Changed

• DeepSeek Provider: Replaced non-functional deepseek-cli with custom provider using official OpenAI client.

Fixed

• Environment Variable Handling: qonqrete.sh now conditionally passes API keys, preventing “unbound variable” errors.

v0.4.8-alpha - 2025-12-06

Added

• DeepSeek Provider: System now supports DeepSeek models via deepseek-cli tool.
• Hybrid Provider Model: AI interaction library supports Python libraries for OpenAI, Gemini, Anthropic, and CLI for DeepSeek.

Changed

• Dynamic API Key Validation: qrane.py reads config.yaml to validate required API keys.

v0.4.7-alpha - 2025-12-06

Added

• Anthropic Claude Provider: System now supports Anthropic’s Claude models.
• Unified Python Provider Interface: Refactored to use official Python clients for all providers.

Removed

• CLI Pre-flight Checks: Removed sgpt and gemini CLI checks as no longer relevant.

v0.4.6-alpha - 2025-12-06

Added

• Configurable Execution Mode: Default mode controlled via cheqpoint boolean in config.yaml.
• --user / -u Flag: Force user-gated mode from command line.

Fixed

• TUI Mode Crash: Fixed “I/O operation on closed file” error.
• Gatekeeper Assessment Parsing: More robust regex-based parsing.
• construqtor Path Duplication: No more nested qodeyard/qodeyard directories.

v0.4.5-alpha - 2025-12-04

Changed

• Logging Architecture: Raw output in struqture/qonsole_<agent>.log, events in struqture/events_<agent>.log.

Fixed

• Headless Mode Crash: Fixed “I/O operation on closed file” error.
• construqtor AI Output Parsing: Extremely strict system prompt with clear output format example.

v0.4.5-alpha - 2025-12-03

Added

• Sqrapyard Project Seeding: On startup, qonqrete.sh checks worqspace/sqrapyard and copies files to qodeyard as a starting point.
• tasq.md Seeding: If tasq.md exists in sqrapyard, it’s used as the initial task.
• Verbose Startup Logging: Explicit logs about seeding or starting fresh.
• Pre-run Delay: 3-second delay after host logs for user readability.

Changed

• Ephemeral Workspaces: Each run creates unique qage_<timestamp> directory for isolation.
• Instruqtor Sensitivity: 10 distinct levels (0-9) via QONQ_SENSITIVITY environment variable.
• Context Awareness: Both instruqtor and construqtor read all qodeyard files for full context.

Fixed

• Stricter Path Sanitization: Removes ../ from AI-generated filenames.
• AI Filename Resilience: Handles AI providing language name instead of filename.

v0.4.4-alpha - 2025-12-02

Fixed

• Instruqtor Logic: Overhauled sensitivity logic.
• Construqtor: Fixed agent failure bug.
• AI Reliability: Robust retry mechanism for intermittent AI failures.

v0.4.3-alpha - 2025-12-02

Added

• Init Seeding: qonqrete.sh init copies sqrapyard to qodeyard if available.

v0.4.2-alpha - 2025-11-28

Added

• Architect Role: Implemented in instruqtor for improved planning.
• Micro-dosing: Technique for better AI results.

v0.4.1-alpha - 2025-11-27

Fixed

• Critical Regressions: Patched syntax errors from v0.4.0.

v0.4.0-alpha - 2025-11-26

Added

• Operational Modes: Agents operate with “personas” via --mode flag.
• Briq Sensitivity: --briq-sensitivity flag (0-9) for task breakdown control.
• TUI Overhaul: Major improvements.

v0.3.0-alpha - 2025-11-25

Changed

• Branding: README.md displays logo.png.
• Versioning: Hardened build process for clean VERSION file.

v0.2.7-alpha - 2025-11-24

• Hotfix: Addressed critical IndentationError in qrane/qrane.py.

v0.2.6-alpha - 2025-11-23

• TUI Experience: Fixed “flash and gone” issue.

v0.2.5-alpha - 2025-11-22

• Agent Stability: Fixed critical NameError and improved console error visibility.

v0.2.4-alpha - 2025-11-21

• Documentation: Consolidated reports into COMING_SOON.md and DOCUMENTATION.md.

v0.2.3-alpha - 2025-11-20

• TUI Stability: Fixed NameError crash in TUI mode.

v0.2.2-alpha - 2025-11-19

Major Refactoring

• Implemented dynamic agent pipeline.
• Centralized path management.
• Added pre-flight checks for dependencies.
• Implemented TUI state persistence.

v0.2.1-alpha - 2025-11-18

• Dynamic Versioning: Centralized in VERSION file.
• Integrated Docker Output: Streamed into TUI.

v0.2.0-alpha - 2025-11-17

Added

• TUI Enhancements: Raw log view, fullscreen mode, key shortcuts.
• Microsandbox (MSB) Integration: Support for msb.

v0.1.1-alpha - 2025-11-14

• TUI Mode: Introduced --tui flag.
• Workspace Cleaning: Added clean command.

v0.1.0-alpha - 2025-11-12

• Initial public alpha release of QonQrete.