179 docs tagged with "scripts"

title: "Add Files To Commits" component_type: script version: "1.0.0" audience: contributor status: stable summary: "CODITECT Git File Staging Helper =======...

title: "Analyze Navigation" component_type: script version: "1.0.0" audience: contributor status: stable summary: "Analyze documentation navigation and cross...

title: "Database path" component_type: script version: "1.0.0" audience: contributor status: stable summary: "Analyze Pilot Launch Work Item Hierarchy Valida...

title: "Get script directory for path resolution (works from any cwd)" component_type: script version: "1.0.0" audience: contributor status: stable summary: ...

title: "Biographical Researcher" component_type: script version: "1.0.0" audience: contributor status: stable summary: "Biographical Researcher - Systematic ...

title: "Configure logging" component_type: script version: "1.0.0" audience: contributor status: stable summary: "CODITECT Interactive Project Setup" keyword...

title: "Custom exceptions" component_type: script version: "1.0.0" audience: contributor status: stable summary: "AZ1.AI CODITECT Master Project Setup Automa...

title: "Configure logging" component_type: script version: "1.0.0" audience: contributor status: stable summary: "AZ1.AI CODITECT Setup Tool" keywords: ['api...

title: "Context Database - Anti-Forgetting Memory System" component_type: script version: "1.0.0" audience: contributor status: stable summary: "Context Data...

title: Intent Context Manager component_type: script version: 1.0.0 audience: contributor status: active summary: Detects context switches, finds matching pl...

title: "Add core scripts to path" component_type: script version: "1.0.0" audience: contributor status: stable summary: "CODITECT Conversation Export Dedupli...

title: "Generate Openapi" component_type: script version: "1.0.0" audience: contributor status: stable summary: "CODITECT OpenAPI Specification Generator ===...

title: "Database path" component_type: script version: "1.0.0" audience: contributor status: stable summary: "CODITECT Learning Management System Query Inter...

title: "Convenience function for quick prompt building" component_type: script version: "1.0.0" audience: contributor status: stable summary: "LLM Abstractio...

title: "=============================================================================" component_type: script version: "1.0.0" audience: contributor status: ...

title: "Add core to path" component_type: script version: "1.0.0" audience: contributor status: stable summary: "Proof-of-Concept: Process Real Claude Export...

title: "Add core to path" component_type: script version: "1.0.0" audience: contributor status: stable summary: "Proof-of-Concept: Same Session Deduplication...

title: "Unified Message Extractor" component_type: script version: "3.0.0" audience: contributor status: stable summary: "Extract ALL data from Claude Code s...

title: "Validate C4 Diagrams" component_type: script version: "1.0.0" audience: contributor status: stable summary: "CODITECT C4 Diagram Validator ==========...

title: "Paths" component_type: script version: "1.0.0" audience: contributor status: stable summary: "Work Items Management - Pop-the-Top Task Automation for...

Script documentation: Adaptive Threshold System for MoE Classification.

Specializes in understanding what makes a document an "ADR" document. ADRs document architectural decisions with context, decision, and consequences. Key ch...

Specializes in understanding what makes a document an "agent" document. Agent docs define AI agent personas, capabilities, and system prompts. Key character...

Intelligent agent selection and invocation system for curriculum development. Analyzes workflows and automatically determines optimal agents, skills, and com...

Generate and manage audit trails for task execution. Supports timeline view, compliance export, and chain verification. Usage: python audit_trail.py --t...

Iteratively classifies documents, injecting content signals until 95-100% confidence is achieved. Forces full signal set at iteration 5 to guarantee 100% cla...

All analyst agents inherit from this base class.

Abstract base class for all judge agents in the MoE classification system. Judges validate analyst votes and can veto classifications that don't meet quality...

Abstract base class for all Type Expert agents. Each expert specializes in one document type and can: 1. Deeply analyze if a document truly belongs to their ...

High-performance batch processing for classifying large document sets. Features: - Resumable processing with checkpoint files - Parallel document processing ...

Script documentation: Batch Corpus Analyzer for MoE Classification.

Implements the staged batch processing pipeline pattern from Agent-Skills-for-Context-Engineering for LLM-powered batch processing. Usage: python3 scrip...

Identifies changelog files - documents that track version history, changes, releases, and updates over time. Key signals: - CHANGELOG in filename - Version ...

Implements a feedback loop that learns from classification outcomes: - Tracks analyst accuracy over time - Dynamically adjusts analyst weights based on perfo...

Hybrid deduplication system for Claude Code conversation exports that combines: - Sequence number tracking (primary deduplication mechanism) - Content hashin...

Part of Track H.2.6: Test agent-to-agent task delegation Based on AUTONOMOUS-AGENT-SYSTEM-DESIGN.md specifications This module provides comprehensive integr...

Comprehensive tests to verify all 130+ agents have correct configuration, valid frontmatter, proper invocation patterns, and can be loaded by invoke-agent.py...

Automatically archives old checkpoints according to checkpoint.config.json settings. Implements never-delete policy with audit trail. Features: - Automatic ...

Implements the Circuit Breaker pattern to prevent cascading failures when calling agents or external services. States: - CLOSED: Normal operation, requests ...

Multi-model AI execution framework - Claude (Anthropic) provider implementation. Provides standardized interface for executing prompts against Claude models ...

Syncs local context data to CODITECT cloud backend. Supports multi-tenant, multi-user, multi-session architecture. Authentication Methods: 1. Email/Pass...

Comprehensive tests to verify all 141 commands have correct configuration, valid frontmatter, proper invocation patterns, and consistent documentation. Run:...

Master test runner that executes all component validation tests and generates a comprehensive gap analysis report. Run: python3 scripts/tests/test_all.p...

This interactive script sets up CODITECT Core on macOS for individual contributors. It handles existing installations, creates backups, and configures Claude...

Shared utility functions for MEMORY-CONTEXT system. Author: AZ1.AI CODITECT Team Sprint: Sprint +1 - MEMORY-CONTEXT Implementation Day 5 Date: 2025-11-16

Autonomous curriculum development lifecycle management system that coordinates multi-level educational content generation, assessment creation, and quality v...

Multi-model AI execution framework - Custom/Local provider implementation. Provides standardized interface for executing prompts against local or custom-host...

Provides unified interface for SQLite (local) and PostgreSQL (production) backends. Automatically selects backend based on environment configuration. Usage:...

Multi-model AI execution framework - Gemini (Google) provider implementation. Provides standardized interface for executing prompts against Gemini models wit...

Multi-model AI execution framework - GPT (OpenAI) provider implementation. Provides standardized interface for executing prompts against GPT models with stre...

Comprehensive tests to verify all 41 hooks have correct configuration, valid structure, and proper documentation. Run: python3 scripts/tests/test_hooks....

Production-grade license management for CODITECT SaaS platform with tier-based feature enforcement, usage quota tracking, and JWT-based license key generatio...

Configures ChromaDB for vector storage and semantic search. Creates collections for sessions and patterns with embeddings. Usage: python3 scripts/core/c...

Backs up SQLite database and ChromaDB vector storage. Supports automated daily backups and restore functionality. Usage: python3 scripts/core/db_backup....

Creates SQLite database with schema for sessions, patterns, and checkpoints. Supports privacy controls, NESTED LEARNING, and context retrieval. Usage: p...

Manages database schema migrations using Alembic. Supports forward and backward migrations for schema versioning. Usage: python3 scripts/core/db_migrate...

Adds sample data to database for testing and demonstration. Creates sample sessions, patterns, tags, and checkpoints. Usage: python3 scripts/core/db_see...

Part of Track H.2: Inter-Agent Communication Infrastructure Based on AUTONOMOUS-AGENT-SYSTEM-DESIGN.md specifications This module provides: - AgentMessage: ...

Part of Track H.2: Inter-Agent Communication Infrastructure This module provides health checking and monitoring for the RabbitMQ message bus: - Connection h...

Extracts reusable patterns from sessions for knowledge accumulation. Implements workflow, decision, and code pattern recognition with knowledge graph constru...

Implements intelligent task routing based on: - Task priority (CRITICAL > HIGH > MEDIUM > LOW) - Agent capabilities (skill matching) - Agent availability (st...

Part of Track H.2.3: Inter-Agent Communication Infrastructure Based on AUTONOMOUS-AGENT-SYSTEM-DESIGN.md specifications This module provides: - PriorityQueu...

Implements privacy-aware data handling with PII detection, redaction, and 4-level privacy model for MEMORY-CONTEXT system. Privacy Levels: - PUBLIC: Can be ...

Integrates privacy_manager.py with checkpoint and export workflows. Provides seamless privacy-aware content processing. Usage: from privacy_integration ...

Part of Track H.2.5: Inter-Agent Communication Infrastructure Based on AUTONOMOUS-AGENT-SYSTEM-DESIGN.md specifications This module provides: - RetryEngine:...

Part of Track H.2.5: Inter-Agent Communication Infrastructure Tests for retry_engine.py - comprehensive coverage of all components Test Categories: 1. E...

Comprehensive tests to verify all 233 scripts have correct syntax, proper documentation headers, and valid structure. Run: python3 scripts/tests/test_sc...

Automatically captures session context for MEMORY-CONTEXT system. Features: - Conversation extraction from checkpoints and sessions - Metadata generation (t...

Comprehensive tests to verify all 186 skills have correct SKILL.md structure, valid configuration, and proper documentation. Run: python3 scripts/tests/...

Part of Track H.2.7: Validate <5s task dispatch latency Based on AUTONOMOUS-AGENT-SYSTEM-DESIGN.md specifications This module validates that the task dispat...

Part of Track H.2: Inter-Agent Communication Infrastructure Based on AUTONOMOUS-AGENT-SYSTEM-DESIGN.md specifications This module provides: - Task: Data mod...

Specializes in understanding what makes a document a "command" document. Command docs define slash commands with invocation syntax, parameters, and usage. K...

Loads components from JSON registries and makes them callable at runtime. Bridges the gap between component files (documentation) and runtime execution. Par...

Provides validated lists of agents, commands, skills, and other components for cross-checking in retrospective analysis and skill optimization. Usage: f...

Ensures ALL modified files are staged for commit, including: - Checkpoint files and indexes - MANIFEST.json files (dashboard dependencies) - Dedup state (glo...

Evaluates context compression quality using probe-based testing from Agent-Skills-for-Context-Engineering. Usage: python3 scripts/context-engineering/co...

Script documentation: Confidence Calibration System for MoE Classification.

Implements the consensus algorithm specified in ADR-019: - Weighted voting based on analyst confidence - Agreement ratio calculation - Threshold-based auto-a...

Validates cross-analyst agreement and ensures classification consistency. Checks for: - Minimum agreement threshold across analysts - Confidence distribution...

Analyzes document classification based on: - Markdown heading structure - Section names and patterns - Code block presence and language - List structures and...

Wraps the degradation detection utilities from Agent-Skills-for-Context-Engineering for CODITECT framework integration. Usage: python3 scripts/context-e...

Wraps the compaction and observation masking utilities from Agent-Skills-for-Context-Engineering for CODITECT framework integration. Usage: python3 scri...

Generate minimal context snapshots (<500 tokens) for efficient session handoffs and state preservation. Usage: python context_snapshot.py python con...

Enhanced with provenance tracking (H.3.4) for full audit trail: - JudgeDecision: model_used, timestamps, token_usage, raw_response - ConsensusResult: provena...

Autonomous project management system for AI curriculum development using the orchestrator agent and educational framework. Creates detailed project plans w...

Implements multi-round debate when judges disagree, based on MAJ-EVAL in-group debate protocol (Chen et al., 2025) and CODITECT research. Key concepts: - MA...

Removes duplicate entries across all major tables: - decisions (by decision text) - code_patterns (by code content) - doc_index (by file_path) - already has ...

Second-tier specialists invoked when first-tier classification is uncertain. These agents perform more expensive but more accurate analysis. Architecture: ...

Script documentation: Directory Judge Validates classification against directory conventions.

Validates classifications against CODITECT domain conventions. Checks for: - Valid document type for CODITECT framework - Classification matches CODITECT nam...

This orchestrator extends the base MoE system with: - Semantic embeddings for improved similarity - Historical learning with dynamic weights - Memory system ...

Ensures all components are: 1. Indexed in SQLite (context.db) 2. Registered in framework-registry.json 3. Activated in component-activation-status.json Part...

=========================================================== Executes tasks using Anthropic Claude (via Claude Code Task tool). Usage: echo '{"task_id":...

================================================== Executes tasks using custom LLM endpoints (OpenAI-compatible or REST API). Usage: echo '{"task_id": ...

=========================================================== Executes tasks using Google Gemini via API. Usage: echo '{"task_id": "TASK-001", ...}' | py...

======================================================== Executes tasks using OpenAI GPT-4 via API. Usage: echo '{"task_id": "TASK-001", ...}' | python...

Maintains consistency between component files (.md, .py, .sh) and JSON registries. Detects drift, auto-updates registries, and validates cross-references. P...

Script documentation: Frontmatter Judge Validates classification against frontmatter declarations.

Script documentation: Generated Task Content Generation

CODITECT Framework - Automated content generation for curriculum tasks Generates multi-level educational content from task specifications with AI-powered con...

Complexity: 3/4 Estimated Duration: 1.1 hours Estimated Tokens: 65,000

Complexity: 3/4 Estimated Duration: 1.1 hours Estimated Tokens: 65,000

Discovers and tracks ALL git repositories under PROJECTS/ directory tree. Handles: - Master repository - All submodules (registered and unregistered) - Stan...

Simple deduplication system that only cares about unique message content. No session tracking - just one global pool of all unique messages ever seen. Key p...

Specializes in understanding what makes a document a "guide" vs other types. Guides are instructional documents that teach users how to accomplish tasks. Ke...

Enables agents and components to register themselves and be discovered by capability. Based on AUTONOMOUS-AGENT-SYSTEM-DESIGN.md specifications. Features: -...

Monitors the component discovery service and reports health metrics. Can be used standalone or integrated into monitoring systems. Usage: python3 script...

Tests cover: - Component registration and unregistration - Capability-based discovery - Type-based discovery - Health filtering and load balancing - Heartbea...

Provides capability-based agent/component discovery by integrating: 1. Component indexer SQLite database (FTS5 search) 2. H.1.5 DiscoveryService (Redis/local...

Tests the capability extraction, matching, and recommendation engine.

Unified metadata registry that consolidates all component metadata sources: - framework-registry.json (component definitions) - capability-registry.json (cap...

Tests the unified metadata registry that consolidates component metadata from multiple sources.

Comprehensive testing of component activation status, file integrity, metadata completeness, and registry consistency. Target: 290+ critical components (age...

Script documentation: Historical Judge Compares against historical classification patterns.

Identifies hook definitions - event-driven automation triggers that execute in response to git events, session events, or other system triggers. Key signals...

Identifies index files - documents that catalog, list, or provide navigation to other documents or components in a collection. Key signals: - INDEX in filen...

Splits large JSONL session files into processable chunks with smart boundary detection and overlap for conversation continuity. Features: - Smart split poin...

Purpose: -------- Process large Claude Code JSONL session files (>10MB) with: - Read-only access (original files never modified) - Streaming chunk processing...

Analyzes Claude Code JSONL session files to identify safe split points, message boundaries, and optimal chunking strategies for large sessions. Features: - ...

Loads and manages judge personas from config/judge-personas.json for the Constitutional Court architecture (H.3.1.7). Features: - Load personas by ID or get...

Provides LLM-backed judge evaluations using the MultiModelClient for multi-provider support with fallback and retry logic. Features: - Uses JudgePersona for...

Ties together all MEMORY-CONTEXT components for end-to-end workflow: 1. Session export from checkpoint 2. Privacy controls (PII detection and redaction) 3. P...

Script documentation: Memory System Integration for MoE Classification.

Merges a remote context.db into the local database with deduplication. Tracks source machine for each record to enable multi-machine sync. Usage: python...

Analyzes document classification based on: - YAML frontmatter fields - Explicit type declarations - Tags and keywords - Status and audience fields

Command-line interface for classifying documents using the Mixture of Experts classification system with 13 Type Experts. Version History: - v2.1: 13 Type E...

Coordinates the full classification pipeline: 1. Run all analysts in parallel 2. Calculate consensus from votes 3. Run judges for validation 4. Produce final...

Provides unified interface to multiple LLM providers for judge evaluations. Supports: Anthropic (Claude), OpenAI (GPT-4), DeepSeek, Alibaba (Qwen), Meta (Lla...

Analyzes document classification based on: - CODITECT naming conventions - Component template patterns - Framework-specific structures - Cross-reference patt...

Generate real-time progress dashboard for PILOT execution plan. Shows track progress, active tasks, blockers, and session statistics. Usage: python pilo...

Verify task completion with evidence-based quality checks. Ensures all requirements are met before marking tasks complete. Usage: python quality_gate.py...

Validates the quality of analyst votes and classifications. Checks for: - Minimum confidence thresholds - Reasoning quality and specificity - Vote completene...

Identifies README files - project introduction and orientation documents that provide overview, setup instructions, and quick navigation. Key signals: - REA...

Specializes in understanding what makes a document a "reference" document. Reference docs provide technical specifications, API documentation, and lookup inf...

Scans CODITECT framework for all components (agents, skills, commands, scripts, hooks, prompts) and populates JSON registries for runtime activation. Part o...

Identifies report files - documents that summarize findings, status, analysis results, or project progress. Key signals: - REPORT/STATUS/SUMMARY in filename...

Maintains a log file with a maximum of 5000 lines. When the limit is reached, the oldest lines (from the head) are removed to make room for new entries. Thi...

Merges generated ADR rubrics with base persona evaluation dimensions and renormalizes weights (H.3.2.4). Features: - Merge ADR-generated rubrics with person...

Analyzes document classification based on: - Natural language understanding of content - Intent classification - Topic extraction - Semantic similarity to do...

Provides true vector embeddings using sentence-transformers for improved document classification accuracy. Replaces regex-based pattern matching with semanti...

Creates and appends to daily session logs (UTC-based), tracks session UUIDs, and auto-classifies after updates. Usage: python3 session_log_manager.py ap...

Tracks processing progress for JSONL session files to enable resume capability after crashes or interruptions. Features: - Per-session watermark tracking - ...

Specializes in understanding what makes a document a "skill" document. Skills define reusable patterns, automation capabilities, and specialized knowledge. ...

Automatically checks for reusable work before executing new tasks to minimize token usage and avoid reinventing the wheel.

Analyzes document classification based on: - File path patterns - Directory location - File extension - File size heuristics

Project: Module 1 Foundations - Priority 1 Content Generation Task ID: TASK_008

Project: Module 1 Foundations - Priority 1 Content Generation Task ID: TASK_004

Project: Module 1 Foundations - Priority 1 Content Generation Task ID: TASK_010

Project: Module 1 Foundations - Priority 1 Content Generation Task ID: TASK_006

Project: Module 1 Foundations - Priority 1 Content Generation Task ID: TASK_011

Project: Module 1 Foundations - Priority 1 Content Generation Task ID: TASK_002

Project: Module 1 Foundations - Priority 1 Content Generation Task ID: TASK_013

Project: Module 1 Foundations - Priority 1 Content Generation Task ID: TASK_007

Project: Module 1 Foundations - Priority 1 Content Generation Task ID: TASK_003

Project: Module 1 Foundations - Priority 1 Content Generation Task ID: TASK_009

Project: Module 1 Foundations - Priority 1 Content Generation Task ID: TASK_005

Project: Module 1 Foundations - Priority 1 Content Generation Task ID: TASK_012

Project: Module 1 Foundations - Priority 1 Content Generation Task ID: TASK_001

Identifies template files - reusable document structures with placeholders that serve as starting points for creating new documents. Key signals: - TEMPLATE...

Verifies that invoke-agent.py can correctly: 1. Load all 130+ agents 2. Parse their frontmatter 3. Generate valid Task() calls Run: python3 scripts/test...

Comprehensive tests for: 1. Persona loading - single and batch 2. Schema validation - required fields and constraints 3. Prompt template rendering - with con...

Tests all 3 judges with sample analyst votes to verify: 1. ConsistencyJudge - agreement validation 2. QualityJudge - vote quality and confidence 3. DomainJud...

Tests the full classification pipeline: 1. Consensus calculation from analyst votes 2. Orchestrator coordination of analysts and judges 3. Parallel execution...

Tests cover: - ADR parsing and frontmatter extraction - Constraint extraction (MUST/SHOULD/MAY patterns) - Technical term detection - Rubric generation logic...

Tests cover: - DebateOrchestrator class - Disagreement detection - Debate context preparation - Debate round execution - Consensus calculator integration

Tests cover: - LLMJudge initialization and configuration - Prompt building - Response parsing - Provenance tracking - LLMJudgePanel multi-model coordination ...

Tests cover: - ModelProvider enum and model routing - CompletionRequest/CompletionResponse dataclasses - FallbackConfig configuration - MultiModelClient init...

Tests cover: - JudgeDecision provenance fields (H.3.4.1) - ConsensusResult provenance chain and dissent (H.3.4.2-H.3.4.4) - ClassificationResult.to_dict() wi...

Tests cover: - EvalCase and EvalResult dataclasses - F1 score computation - Eval runner with mocked API calls - Critic agent response parsing - Improvement a...

Orchestrates all Type Expert agents to achieve consensus on document classification. This is the core of Phase 2 - coordinating expert analysis, vote reconci...

Automatically detects environment and configures appropriate logging backend: - LOCAL: File-based logging with RollingLineFileHandler - GCP: Cloud Logging wi...

Tests cover: - CircuitBreaker state machine (CLOSED → OPEN → HALF_OPEN → CLOSED) - Failure counting and threshold transitions - Recovery timeout behavior - C...

Part of Track H.2: Inter-Agent Communication Infrastructure Tests cover: - AgentMessage serialization/deserialization - MessageBusConfig from various source...

Tests cover: - Queue configuration and management - Routing rules and task routing - Dequeue strategies (strict, weighted, round-robin, fair-share) - Priorit...

Tests: - AgentStatus enum - AgentCapability dataclass - Agent dataclass - LocalAgentRegistry - AgentDiscoveryService - PriorityRouter (all strategies) - Inte...

Part of Track H.2: Inter-Agent Communication Infrastructure Tests cover: - Task data model serialization/deserialization - TaskQueueConfig from various sour...

ONE script to invoke ANY of the 130+ CODITECT agents without needing individual wrappers, skills, or complex dispatchers. Usage: python3 invoke-agent.py...

Intelligently leverages previously created work to minimize token usage and avoid reinventing the wheel. Scans existing content, scripts, and patterns for re...

Specializes in understanding what makes a document a "workflow" document. Workflows define processes, pipelines, and multi-phase operations. Key characteris...