Pseudolife-MCP

# Pseudolife-MCP

> Persistent long-term memory for Claude Code, Codex, and other MCP clients: an associative store that ages and supersedes, a slot-keyed canonical-fact cortex, dream consolidation into facts and a knowledge graph, procedural lessons, and cited world facts — served by a local daemon your coding agent calls over MCP.

## Documentation

- [Pseudolife-MCP](https://raw.githubusercontent.com/Pseudogiant-xr/Pseudolife-MCP/master/README.md): Overview, quickstart, tool reference, and operations — start here
- [The memory model — cortex, world facts, lessons, and time](https://raw.githubusercontent.com/Pseudogiant-xr/Pseudolife-MCP/master/docs/guide/memory-model.md): The canonical-fact layers: cortex slots, supersession, contenders, world facts, lessons, and the temporal/multi-writer stamp
- [Retrieval — search, recall, and the knowledge graph](https://raw.githubusercontent.com/Pseudogiant-xr/Pseudolife-MCP/master/docs/guide/retrieval.md): How memory_search ranks: hybrid dense+lexical scoring, reranking, and the explain trace
- [Dreaming — consolidating memories into facts](https://raw.githubusercontent.com/Pseudogiant-xr/Pseudolife-MCP/master/docs/guide/dreaming.md): The consolidation pass: extractors, the cursor, claim gating, deep dream, and the review queue
- [Episodes & session lifecycle](https://raw.githubusercontent.com/Pseudogiant-xr/Pseudolife-MCP/master/docs/guide/episodes.md): Session-scoped episodes: attribution, auto-titles, resume, and consolidation
- [Configuration](https://raw.githubusercontent.com/Pseudogiant-xr/Pseudolife-MCP/master/docs/guide/configuration.md): Every configuration knob, the DSN, schema version history, and extractor setups
- [Recovering agent mailboxes after a database restore](https://raw.githubusercontent.com/Pseudogiant-xr/Pseudolife-MCP/master/docs/guide/coordination-recovery.md): Offline mailbox recovery after a full restore: revoke attachments, rebind private adapter state, and explicitly resume delivery
- [Providers — one memory bank across every coding agent](https://raw.githubusercontent.com/Pseudogiant-xr/Pseudolife-MCP/master/docs/guide/providers.md): Per-coding-agent capability matrix, the hook-equivalent ladder, the AGENTS.md standard, and writer ids
- [Benchmarks](https://raw.githubusercontent.com/Pseudogiant-xr/Pseudolife-MCP/master/docs/guide/benchmarks.md): LongMemEval methodology and results, with the committed evidence artifacts behind each number
- [Comparison — where this sits among agent-memory projects](https://raw.githubusercontent.com/Pseudogiant-xr/Pseudolife-MCP/master/docs/guide/comparison.md): Where this sits among agent-memory projects, the axes it is built around, and when to use something else instead
- [Security posture — memory poisoning (ASI06)](https://raw.githubusercontent.com/Pseudogiant-xr/Pseudolife-MCP/master/docs/guide/security-posture.md): Memory poisoning (ASI06): the threat model, every shipped mitigation mapped to it, and what is not defended

## Optional

- [CHANGELOG](https://raw.githubusercontent.com/Pseudogiant-xr/Pseudolife-MCP/master/CHANGELOG.md): dated, versioned change history in Keep-a-Changelog format