This Tool Gives Your Coding AI a Memory That Lasts Between Sessions
A new open-source tool called funes lets AI coding assistants remember past decisions, failed experiments, and the reasoning behind your code, even after you close the session and reopen it days later.
Key points
- funes is a free, open-source memory layer for AI coding assistants including Claude Code, Codex, pi, and Hermes, announced via Hugging Face.
- All indexing and search runs locally on your own machine; no cloud service processes your code or sessions.
- A shared memory option lets teams share months of past decisions with a new teammate on their very first day.
- In a small benchmark test, recall via funes was eight times cheaper than the leading alternative for retrieving prior session knowledge.
- The tool publishes shared memories as private Hugging Face datasets by default, with credentials scrubbed before anything leaves your machine.
Every time you start a new session with an AI coding assistant, the same thing happens: the AI meets your project as if it has never seen it before. Ask it why the team switched away from the streaming parser three weeks ago, and it has no idea. That history simply vanished when the last session ended.
funes, a new open-source tool announced through Hugging Face, tries to fix that. It sits quietly alongside the AI coding tools you already use and builds a searchable record of every decision, dead end, and piece of reasoning those tools have produced.
How does it actually work?
Simply put: funes watches your AI coding sessions and files them away in a searchable index on your own computer. When a new session touches a topic covered in an old one, the AI can reach back into that index and retrieve the relevant passage, including the exact session, timestamp, and turn it came from.
Setup takes two commands. You install funes with a script, then run one command to attach it to your chosen coding tool. After that, you work normally. The memory builds itself in the background.
The retrieval system combines two search methods: vector search (which finds passages that are conceptually similar to your question) and BM25 search (a keyword-matching approach familiar from search engines). A second AI model, called a cross-encoder, re-ranks the results to surface the most relevant ones. All of this runs on your machine. No external service sees your code.
What does it return when you ask it something?
The full original text, not a summary. That distinction matters. When an AI compacts a long session into a shorter summary, it inevitably drops detail. funes returns the actual passage, so nothing has to survive a summarisation step.
You can also query a memory directly from the command line without involving an AI agent at all: funes ask claude "what did we decide about the streaming parser". The tool retrieves relevant passages, hands them to the AI, and returns an answer that names its sources. If the passages do not support an answer, it says so rather than making one up.
A small benchmark comparing three approaches to long-session recall found recall via funes cost roughly eight times less than the next option, and it was the only approach that succeeded on both test tasks. Compaction, the method most AI agents use by default, failed on one of the two tasks because its summary had dropped the findings that mattered.
Can a whole team share the same memory?
Yes, and this is where funes becomes interesting for organisations. Bind a funes memory to a Hugging Face dataset you own (private by default) and it follows you across machines. Any teammate who connects to the same dataset can search months of past decisions, including rationale that never made it into a pull request or a document.
Before anything leaves your machine, funes scrubs credentials from the data. It then scans again during the publishing step. The project documents what the scanner covers and where its limits are.
Open-source projects can also publish a memory alongside a release, giving contributors a searchable record of why the project works the way it does.
Common questions
Does funes send my code to any outside server?
No. Indexing, embedding (converting text into a searchable numerical form), and re-ranking all run locally. A shared memory publishes to a Hugging Face dataset you own and control, not to a third-party service.
Which AI coding tools does it support?
funes currently supports Claude Code, Codex, pi, and Hermes. The tool is open source, so support for other agents may follow.
What happens if funes cannot find a relevant answer?
It says so. The tool does not invent an answer when the retrieved passages do not support one. You can rephrase your question or let the AI agent search the memory more deeply during a normal work session.
