MEMANTO — Persistent Memory for AI Agents

MEMANTO — Persistent Memory for AI Agents

MEMANTO is a universal memory layer that gives AI agents long-term, persistent memory across sessions. Built on Moorcheh's semantic database, it provides zero-cost ingestion latency, instant recall, and a rich taxonomy of memory types with trust scoring.

Key Capabilities

Agent & Session Management allows creating named agent identities, activating sessions with JWT tokens, maintaining continuity across restarts, and deleting agents. Deletion removes local metadata and prompts whether to also purge the Moorcheh cloud namespace.

File Upload ingests documents (.pdf, .docx, .xlsx, .json, .txt, .csv, .md) directly into the agent's memory namespace. Uploaded content is processed, embedded, and immediately searchable via recall and answer.

Memory Storage supports 13 distinct memory types — from facts and decisions to errors and commitments — each with confidence scoring and provenance tracking to prevent memory poisoning.

Semantic Recall enables querying all stored memories using natural language, with filtering by type, confidence threshold, time range, and tags.

RAG Answers generates grounded natural-language responses by retrieving relevant memories and synthesizing them into coherent answers.

Memory Sync exports the full memory snapshot to MEMORY.md in the project root, ensuring agents always start with full context.

Daily Summaries auto-generates compressed digests of recent activity, reducing context overhead for long-running agents.

Getting Started

Users need a Moorcheh account with an API key. Create a free account at console.moorcheh.ai.

See environment_requirements.md for full setup instructions.

First-Time Setup

pip install memanto
export MOORCHEH_API_KEY="your-api-key"
memanto
memanto agent create my-agent
memanto agent activate my-agent

Session Start Pattern

memanto memory sync --project-dir .   # Populate MEMORY.md
memanto recall "instructions decisions goals" --limit 20
memanto answer "What are my pending commitments?"

Usage Notes

Memory Types Decision Matrix:

Type	When to Use	Confidence	Example
fact	Verified information, project status	0.9–1.0	"API uses PostgreSQL for metadata"
decision	Architecture choices, approach selections	0.9–1.0	"Chose React over Vue for frontend"
instruction	Standing rules, preferences, guidelines	0.9–1.0	"Always use type hints in Python"
commitment	Promises, TODOs, obligations	1.0	"Will deploy monitoring by Friday"
preference	User/team preferences	0.8–1.0	"User prefers dark mode"
goal	Objectives, targets, milestones	0.8–1.0	"Launch CLI by end of March"
artifact	Tool outputs, reports, file locations	0.9–1.0	"Report saved at ./reports/q1.md"
learning	Knowledge acquired from experience	0.7–0.9	"Batch operations 100x faster"
event	Important conversations, milestones	0.8–0.95	"Completed Phase 1 features"
relationship	Team context, collaboration patterns	0.85–0.95	"Alice is lead backend engineer"
observation	Patterns noticed, behaviors	0.6–0.85	"User prefers short responses"
error	Failures, bugs, lessons learned	0.95–1.0	"Namespace format bug — use underscores"
context	Session summaries, status updates	0.9–1.0	"Project 70% done, API complete"

Confidence Levels:

• 1.0 — Explicit user statement, verified fact, standing instruction
• 0.9–0.95 — Strong consensus, well-tested approach
• 0.8–0.85 — Observed pattern (3+ times)
• 0.7–0.75 — Emerging pattern (2 times), reasonable inference
• 0.6–0.65 — Single observation, uncertain
• < 0.6 — Do NOT store. Too uncertain.

Provenance Types: explicit_statement · inferred · observed · corrected · validated · imported

Common errors relate to missing API keys, expired sessions (re-run memanto agent activate), and missing agent IDs. See individual reference files for details.