Roampal gives agent tools a private memory layer for chats, preferences, and outcomes. It uses a multi-tier memory system, semantic retrieval, and feedback-based scoring so the stored context improves over time.
Builders who want Claude Desktop, Cursor, or another MCP tool to remember what happened in earlier chats.
You can stop re-explaining yourself and let your agent reuse remembered context, preferences, and outcomes.
Scores replies and memories so good results are promoted and bad ones can be removed or downranked.
Separates Books, Working memory, History, and permanent Memory Bank data so different kinds of context live at different timescales.
Exposes tools like `search_memory`, `add_to_memory_bank`, `update_memory`, `archive_memory`, `get_context_insights`, `record_response`, and `score_memories`.
Keeps data on your machine and works offline after model download.
Lets you upload `.txt` and `.md` files as searchable reference material.
Works with tool-calling models through Ollama or LM Studio.
Settings → Integrations → Connect → Restart your tool
Memory that learns what works. So you can do more of it.
Say it worked. Say it didn't. The AI remembers.
Stop re-explaining yourself every conversation. Roampal remembers outcomes, learns from feedback, and gets smarter over time—all 100% private and local.
85.8% non-adversarial on LoCoMo (1,986 questions). +23 pts over raw ingestion. Absorbs 1,135 poison memories losing only 4 pts. (Paper)
LoCoMo dataset (1,986 questions, 5 categories, corrected ground truths). Evaluated with roampal-labs. Dual-graded by local 20B + MiniMax M2.7.
| Metric | Result |
|---|---|
| Non-adversarial accuracy (MiniMax-regraded) | 85.8% |
| Overall (all 5 categories) | 76.6% |
| vs raw ingestion baseline | +23 pts (76.6% vs 53.0%, p<0.0001) |
| Poison resilience | -4.2 pts after 1,135 adversarial memories |
| No-memory baseline | 6.0% (model has zero LoCoMo knowledge) |
| Architecture vs model | Architecture: +23 pts. Model swap (GPT-4o-mini): 1.5-2.5 pts |
| Config | Hit@1 Clean | Hit@1 Poison | p-value |
|---|---|---|---|
| TagCascade + cosine | 27.3% | 29.0% | baseline |
| Overlap + cosine | 25.8% | 28.0% | p=0.0003 |
| Pure CE | 25.4% | 28.4% | — |
| TagCascade + Wilson | 23.0% | 25.0% | p<0.0001 |
Full methodology in roampal-labs
Roampal.exe → Run as administratorYour AI starts learning about you immediately.
Memory That Learns
Your Knowledge Base
Privacy First
Connect Roampal to Claude Desktop, Cursor, and other MCP-compatible tools.
Settings → Integrations → Connect → Restart your tool
7 tools available: search_memory, add_to_memory_bank, update_memory, archive_memory, get_context_insights, record_response, score_memories
┌─────────────────────────────────────────────────────────┐
│ 5-TIER MEMORY │
├─────────────┬─────────────┬─────────────┬──────────────┤
│ Books │ Working │ History │ Patterns │
│ (permanent) │ (24h) │ (30 days) │ (permanent) │
├─────────────┴─────────────┴─────────────┴──────────────┤
│ Memory Bank │
│ (permanent user identity/prefs) │
└─────────────────────────────────────────────────────────┘
Core Technology:
Works with any tool-calling model via Ollama or LM Studio:
| Model | Provider | Parameters |
|---|---|---|
| Llama 3.x | Meta | 3B - 70B |
| Qwen 2.5 | Alibaba | 3B - 72B |
| Mistral/Mixtral | Mistral AI | 7B - 8x22B |
| GPT-OSS | OpenAI (Apache 2.0) | 20B - 120B |
| Document | Description |
|---|---|
| Architecture | 5-tier memory, retrieval pipeline, technical deep-dive |
| Benchmarks | LoCoMo evaluation, TagCascade results |
| Release Notes | Latest (v0.3.3): multimodal image input, dynamic capability + context detection, ChromaDB phantom-handling closing issue #8, atomic config writes, Harmony token cleanup |
AI Safety: LLMs may generate incorrect information. Always verify critical information. Don't rely on AI for medical, legal, or financial advice.
Model Licenses: Downloaded models (Llama, Qwen, etc.) have their own licenses. Review before commercial use.
Free & open-source (Apache 2.0 License)
Made with love for people who want AI that actually remembers
Sign in to join the discussion.
No comments yet. Be the first to say what this is good for.
High-performance code intelligence MCP server. Indexes codebases into a persistent knowledge graph — average repo in milliseconds. 158 languages, sub-ms queries, 99% fewer tokens. Single static binary, zero dependencies.
Universal provider proxy for OpenAI Codex & Claude Code — use any LLM (Claude, Gemini, Grok, DeepSeek, Ollama…) with Codex CLI, App, SDK, and Claude Code
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.
Local-first code intelligence graph for MCP and CLI. Builds a persistent map of your codebase so AI coding tools read only what matters, with benchmarked context reductions on reviews and large-repo workflows.
Stop your AI from making things up — it proposes, deterministic tools decide, every claim checked against ground truth with evidence. Grounded facts and context survive resets. Reverse engineering is the proving ground. MCP server + CLI.
20 MB lightweight cross-platform database client for 90+ databases, including MySQL, PostgreSQL, SQLite, Redis, MongoDB, DuckDB, SQL Server, and Dameng. Built-in AI, MCP Server, CLI, desktop and Docker. | 轻量级跨平台数据库管理工具，支持 MySQL、PostgreSQL、SQLite、Redis、MongoDB、达梦等 90+ 数据库，提供桌面端、Docker、CLI、内置 AI 助手和 MCP Server。