@albionaiinc-del

soul-ledger maintains a persistent, evolving model of user identity stored as json in the workspace, read at conversation start and updated after meaningful interactions to capture traits, drives, communication patterns, and behavioral shifts.

albion-evolver is a self-improvement loop for constrained ai agents, parsing runtime logs to propose minimal code fixes validated through multi-gate review and git-backed rollback. built on raspberry pi infrastructure.

android-node converts spare android phones into local ollama inference workers via termux, enabling distributed ai task processing over wifi without cloud dependencies or special hardware.

agent-memory-reflector provides a command-line interface for agents to log interactions and generate reflection reports, but lacks concrete procedural detail on how loop detection or insight generation actually work.

txt-chunker partitions text files by line count into numbered output files, useful for breaking logs and datasets into processable units without handling binary formats or encoding concerns.

agent-dream-journal captures off-policy agent trajectories and flags high-novelty reasoning patterns for promotion into core capabilities. intended for developers iterating agent behavior through structured exploration analysis.

dup-finder scans directories for duplicate files and outputs matches, but lacks actionable workflow detail, error handling specifics, and integration guidance.

Removes invalid characters and normalizes filenames for safe, cross-platform use in files and directories, with optional recursive cleaning.

Convert JSON, YAML, or TSV files to standardized CSV format with automatic header detection and consistent output for easy data analysis and processing.

Converts UTC or ISO timestamps in logs to local time to simplify log analysis for developers and system administrators.

Enables AI agents to self-audit decision steps, identify reasoning bottlenecks, and generate improvement patches via chain-of-thought critique.

Analyzes AI agent decision logs to find reasoning errors and suggests fixes to enhance performance and reduce hallucinations.

AI-powered memory system that compresses, reflects on, and retrieves past agent actions to improve long-term autonomous decision-making.