fava-trails
Active·★ 26·Apache-2.0·Updated 2026-10-21
FAVA Trails is a Git-native system providing a versioned audit trail and curated memory for AI agents, storing thoughts, decisions, and observations as markdown files in a controlled Git repository. It ensures crash-proof persistence, full lineage tracking, and integrates with LLMs for trust gate verification and context engineering through the Model Context Protocol (MCP).
#AI Agents#Knowledge Base#Version Control#Audit Trail#Memory Management#LLM Integration#Context Management#Lifecycle Hooks
01
Features
01Supersession tracking for consistent memories, hiding outdated beliefs from recall.
02LLM-based Trust Gate for validating thoughts and containing hallucinations in drafts.
03Full lineage tracking and crash-proof persistence with atomic Git commits.
04Draft isolation, ensuring only promoted thoughts are visible to other agents.
02
Compatibility
Python
Runtime
Verified via docs
Git
VCS
Verified via docs
MCP (Model Context Protocol)
Protocol
Verified via docs
OpenRouter
LLM Provider
Verified via docs
03
Quick start
1
$ pip install fava-trails
04
Use cases
↳Persistent and versioned storage for AI agent thoughts, decisions, and observations.
↳Facilitating collaborative AI agent workflows with shared, validated knowledge bases.
↳Auditing and debugging AI agent behavior through a complete, versioned history.
↳Enhancing LLM context management with protocols like SECOM for compression and ACE for reranking.
05
Alternatives
agents-remember★ 29
`Agents Remember` is a Drift-aware repository memory for coding agents in complex codebases. Captures what code can't say on its own! Retrieves memory by path, semantic search, and relationship (code-graph).
okf-agent-memory★ 827
Git-native persistent memory for AI coding agents. Implements Google OKF v0.2 with sub-300µs in-memory BM25 search, embedded MCP server, and progressive disclosure. Slashes token bloat by 80% with zero external databases or dependencies. Built in pure Go.
mcp-local-memory★ 60
A lightweight, powerful local memory server for AI agents supporting text, entities, and relations. Enables persistent codebase understanding and user preference management.
mori★ 32
mori MCP server — A shared memory layer for AI coding agents — one that compounds.
Comments
Log in to leave a comment
No comments yet. Be the first!