memoket/memoket-kite

Most AI memory turns every sentence into a cloud of numbers and hands back whatever lands closest to your question. It always returns something, even when the answer was never there, and it cannot show you why.

KITE by Memoket reads what a question means, not how it sounds.

"Who's the decision-maker on the Henderson account now?"

Nothing in memory says decision-maker:

• March. Marcus: "I'll sign off on Henderson."
• June. The new VP of Ops: "Dana owns Henderson from here."

Both were filed under the same topic when they were written. Then now sorts them by date. Similarity has no notion of now: nearest-match memory rates March just as relevant and answers "Marcus", with nothing to tell you it is wrong. KITE answers "Dana", with the June line attached.

The whole idea in one line: conversations settle into facts under topics, and a question finds its topic and pulls that topic's memories, receipts included.

A user says, "I moved to Tokyo on July 26." KITE by Memoket extracts what happened and files it under its topics, keeping the original message as evidence:

<fact t="2026-07-26" kind="event" who="user"
      place="tokyo" event="move" src="chat1L1">
  The user moved to Tokyo on 26 July 2026.
</fact>
<line id="chat1L1" who="user">I moved to Tokyo on July 26.</line>

When the agent asks "Where does the user live now?", KITE by Memoket compiles the question into a plan you can read, validate, and cache:

{
  "select": "facts",
  "where": {"who": ["user"], "event": ["move"]},
  "pipe": [
    {"op": "sort", "key": "t", "desc": true},
    {"op": "head", "n": 1}
  ]
}

The executor runs the plan over the topic index. Keywords and associations come first, then conditions filter down to the one moment that answers the question:

SEARCH facts
FILTER who = "user" AND event = "move"
SORT BY event_time DESC
TAKE 1

The same validated plan always follows the same steps over the same memory, with no embeddings and no vector database anywhere.

Answer: Tokyo
Source: chat1L1, "I moved to Tokyo on July 26."

python -m pip install memoket-kite
export OPENAI_API_KEY="sk-..."
mkdir -p artifacts
curl -fsSL \
  https://raw.githubusercontent.com/memoket/memoket-kite/main/examples/data/demo_codebook.xml \
  -o artifacts/quickstart.xml

Using another OpenAI-compatible provider? Set OPENAI_BASE_URL as well.

Turn a conversation into structured, source-backed facts:

from memoket_kite import Memory

memory = Memory.load("artifacts/quickstart.xml")
memory.remember(
    [{"role": "user", "content": "I moved to Tokyo."}],
    session_id="session_1",
)

Compile a question into a plan and retrieve matching facts, each carrying its sources:

from memoket_kite import Memory

memory = Memory.load("artifacts/quickstart.xml")

for fact in memory.recall("Where did the user move?"):
    print(fact.content)
    print(fact.sources)

Generate an answer from the retrieved evidence and keep the receipts:

from memoket_kite import Memory

memory = Memory.load("artifacts/quickstart.xml")

result = memory.answer_with_evidence("Where did the user move?")
print(result.text)  # "The user moved to Tokyo."
for fact in result.evidence:
    print(fact.content, fact.sources)  # the moments the answer stands on

print(memory.answer("Where did the user move?"))  # just the text

Full reference in docs/api.md; more runnable examples in examples/.

KITE by Memoket posts the top overall score on both public long-conversation memory benchmarks. It is the only system up there that uses no vectors, and it reads less context than any capable rival.

All systems are evaluated under one shared protocol, with gpt-4.1-mini as the common reader and judge; the exact rubrics are in the LoCoMo protocol and LongMemEval protocol. Reader context is the mean per-question evidence budget: what the reader actually gets to see, measured by the same ledger on every question.

Each figure comes from a reference run under the default configuration, with gpt-4.1-mini reading and judging and no ablation flags set. The reproduction contract pins the dataset revisions, protocols, model IDs, expected row counts, and reported scores. The provided scripts rebuild the evaluation locally from the official datasets with a configured compatible LLM endpoint.

KITE by Memoket is a plain Python library: three calls take you from a raw conversation to a cited answer. Integrations across the agent ecosystem (Claude Code, Codex, Cursor, OpenCode, and more) are next on the roadmap.

Want yours first? Tell us on Discord.

The memory you just read about ships today. KITE is the algorithm inside every Memoket: the same topic-filed facts, the same line back to the source, running on your own device. Ask it "didn't this client say their fiscal year ends in March?" months later, and it surfaces the exact moment: who said it, when, and in their words.

Memoket captures memories; KITE is the memory itself.

See more of Memoket in action →

• 💬 Discord for questions, ideas, and integration requests
• 🐛 Issues for bugs and feature requests
• 🔧 Contributing guide · 🔒 Security policy

A technical report is in preparation; BibTeX will land here with the arXiv release. Until then, please cite this repository.

KITE by Memoket is released under the Apache License 2.0.

Like its name, KITE by Memoket stays light enough to fly, always on a line back to the truth. 🪁