KRouter
io.github.398894496-archPublished by GitHub account 398894496-archv2.1.0Updated Nov 11, 2026
Plain-Markdown memory for your AI: every answer cites its page, and it says when it does not know.
The publisher's own server: 370.ai — AI Gateway: 100+ Chat Models, Image & Speech Generation
Overview
A local Markdown memory for coding agents: it recalls what was settled before, cites the page, and says when it does not know.
- What you can do
- Ask the memory a question and get the matching page, a one-line answer and whether it refused
- Write a single line down as a note (a decision, a correction, a fact) on that day's notes page
- Check the memory's status and see what it holds and why a lookup answered as it did
- Search a folder of existing Markdown as a knowledge base, with reference pages listed apart from your own decisions
- Import PDF, Word or HTML files into the knowledge base as Markdown pages
- When to use it
- Worth adding when you want a coding agent to remember earlier decisions, rules and corrections across sessions without a cloud service or API key, and to show which page each answer came from. Also useful if you already keep notes in Markdown and want them searchable alongside your decisions.
- Requirements
- Runs as a local process over stdio, with standard-library Python and no key or server for one person. The installer wires the AI apps it finds; other MCP apps need the server added to their settings with a full path to the Python script. Importing non-Markdown files needs a converter such as markitdown, pandoc or pdftotext.
Installation
In SourceWeft
- Open KRouter in the dashboard and add it to a workspace.
- Enable the server for the chats that should use its tools.
Desktop only via STDIO. STDIO servers start a local process, so they need the SourceWeft desktop host.
Other MCP clients
Follow the launch instructions in the repository.
README
KRouter
Give your agent a subconscious: mention it and it comes to mind, with what is connected to it, and the lessons carry over.
"Subconscious" is a figure of speech: recall that happens without you asking. A lookup on your machine, no model, nothing sent.
KRouter is a local Markdown memory for coding agents (Claude Code, Codex, Cursor, DeepSeek Harness, Hermes, OpenClaw) and for any app that speaks MCP, Claude Desktop for one. When something you mention was settled before, the agent recalls it, with what is connected to it; a lesson from one task is offered on another; a change of mind replaces the old answer; every answer names the page it came from. For one person it needs no API key and no server; a team server and model distilling are optional (below).
You: Going to move the clippings originals into the archive folder this afternoon.
Agent: That note says not to move the Clippings originals (from your "User preferences" page under 02 Methods/Agent), so I'd hold off on moving them to the archive this afternoon. If you meant a copy or a different set of files, tell me and we can sort out a safe way to do it.
You: The coffee machine on our floor is broken again.
Agent: Sorry to hear that, it's annoying when it keeps happening. I can't fix it from here, but I can help you draft a quick message to facilities if you'd like.
You only mentioned it in passing; KRouter recognised something already settled in what you said and handed it to the agent with the page it came from. The second remark touches nothing in the memory, so nothing was added and the agent answered as usual. Both agent replies are real: Claude Sonnet 5.5 (a Claude Code subagent, no tools), 2026-11-10, given exactly what the collapsed block below shows, copied as it answered; with another agent, the reply is up to that agent.
What the agent was actually handed (the prompt hook's raw output on the English sample memory)
Install
It asks nothing and connects the AI apps it finds. An existing folder keeps every file; only missing ones are added,
and uninstalling never touches your memory. Windows, the Claude Code plugin, --minimal, every file it changes, what
works without a key, and uninstalling: docs/INSTALL.md.
See it in a few seconds, nothing installed, no key: the same story on a throwaway memory, deleted afterwards.
Your own vault, Claude Desktop and Obsidian: a writer's example.
Mention it and it comes to mind
Every message is looked up locally, and the session starts with an index of what has been settled, so the agent also catches a thing said in other words. Held out, 24 mentions (two thirds without the page's names): Sonnet 5.5 24 / 24, Haiku 5.5 23 / 24; messages that should change nothing moved it 1 / 24 and 2 / 24 (set 35).
With what is connected to it
A lookup gives five pages: three that match, then up to two the memory connects to them (a link, a replacement, the same session, rare words they share, being cited together). Held-out half of LongMemEval, five pages each: every session the answer needs for 210 of 235 questions, keyword search 198; multi-session 40 against 31 (set 18).
The lessons carry over
A lesson from one task is offered on another in the same situation, and lessons that keep coming back grow into
skills the agent fetches when they fit. Held out: the right lesson for 14 / 14 (Sonnet 5.5) and 11 / 14 (Haiku
5.5) tasks with a new subject, 14 / 14 and 13 / 14 in English (sets 15, 16); skills drafted right 16 / 16 and
fetched and followed 12 / 12 (sets 24, 25; bench/memory_honesty/HELDOUT.md).
Underneath, what you can check
Every answer comes with the page it came from; when your memory has nothing on a question, the receipt says so
outright, when it has only something close it says how sure it is instead of passing that off as the answer, and
reference material you keep never stands in for what you decided; a correction replaces the old answer instead of competing with it; and you can see all of
it: the memory is a folder of plain Markdown you can open and edit, in Obsidian or any editor,
and krouter view / krouter why show what it holds and why it answered. It works with the AI you already use, and
one is enough.
Refusing comes in two layers, and only the first is KRouter's. On LongMemEval's 30 unanswerable questions KRouter's
own receipt says recall: none for 9. With a reader model on top, 29 of the 30 were answered "not mentioned", but the
same reader given nothing at all said so for all 30, so that second number is mostly the model's doing. What KRouter does:
when nothing in the memory matches, it says so outright; when only something close matches, it hands the model the page
and a confidence level instead of passing a near match off as the answer. Not a guarantee without exceptions: for 22 of
those 30 the best page is a real session about something near what was asked. Reference material you declare as
your knowledge base never counts as the memory's answer: when only it matches, the
receipt's first line is memory: none. A rule to refuse more near matches was looked for on development data (set 37)
and none was good enough, so the numbers above stand. Instead the receipt's first line says which case it is:
memory: direct, memory: related (only related pages: answer only from a sentence that states it outright, else
say the memory does not have it) or memory: none; for the 30, 0 / 21 / 9. Strict mode (KROUTER_STRICT=1, off by
default) treats related as none; on a memory of session pages that means it answers almost nothing (set 38). Method and the other benches: docs/NUMBERS.md.
What it does not do
- The samples are small: a few dozen cases per set.
- In these evaluations the host wrote down which page it would fetch; it did not call the tools for real.
- In real sessions hosts almost never write a lesson card on their own (0 of 4 on both hosts, sets 30 and 31): learning comes from the hooks and the nightly run.
- On Cursor the per-message layer is not available (its hook cannot add context); it gets the session-start index.
- Answer accuracy has not been rerun for the five-page version (set 22 is registered, not run).
- Windows was not tested for 2.0.
- A question about something near what your memory holds still gets an answer, usually at
lowormedium. Paragraph evidence (set 39) refuses more on LoCoMo (18 → 61 of 223) but not on LongMemEval (9 → 10 of 30): off by default. - Reference material is recognised page by page without declaring (set 40: 30 of 32 held-out articles, 0 of 43 of
your own pages taken for one), on a synthetic vault only;
krouter kb mine PAGEcorrects a wrong call.
[ci] [release] [license] · 中文：README.zh.md
Engineering facts, each one checkable now: created 2026-08, one maintainer so far (MAINTAINERS.md).
- Every push runs the full suite (687 tests as of 2.1.0;
python3 -m pytest -q) on Linux, macOS and Windows, and a container smoke test (ci.yml). - Each release carries SHA-256 sums and a GitHub build-provenance attestation (
gh attestation verify <package> --repo 398894496-arch/KRouter); the one-line install uses a tag, nevermain, and checks the sum before running anything. - After a release, a job runs every install command this README gives against the published addresses (
release_smoke.py,release.yml); the container image isghcr.io/398894496-arch/krouter. - In the official MCP Registry as
io.github.398894496-arch/krouter. - Evaluations are registered before they are run (rows that were not say so in
NUMBERS.md); question sets and scripts are in the repo (HELDOUT.md, rerun them). SECURITY.md,CHANGELOG.md,./scripts/install.sh --upgrade,./scripts/uninstall.sh, and any earlier release withKROUTER_TAG=(DETAILS.md). What you ask never leaves your machine.
No key is needed to start: lookup, sources and refusals run with no model, and the newer page of a changed decision is listed first and flagged (what works without a model).
For a team
Tested on one machine only; no real team has used it over weeks. One server holds your team's memory: a folder of Markdown in a git repository, run by you (there is no KRouter cloud).
Each person gets their own token, every write is recorded under the writer's name, and "who changed this, and when" can be
asked (krouter_changes); when two people correct the same decision, the last to arrive is in force and the earlier one
is kept as replaced, named in the reply. Ten people's agents on one laptop (set 27): 0 errors in 440 calls at each of
three sizes, all 204 writes per size under the right person's name, none lost; 30 minutes at 10,136 pages: 21,218
calls, 0 errors. Deployment, tokens, TLS and the
numbers: docs/TEAM.md.
It is also your knowledge base
Point KRouter at a folder of Markdown you already have (an Obsidian vault, a docs repository, a notes directory) and every page in it can be found with the page it came from. The folder holds two kinds of page: your memory (what you decided, your rules and corrections, what sessions were distilled into) and your knowledge base (reference material: articles, clippings, research, documents someone else wrote). Both are searched; only your memory counts as what you decided.
Clipping folders (Clippings, 剪藏, Web clips) and pages with a web clipper's frontmatter count without being
declared. A lookup lists knowledge-base pages in a block of their own, each with its title, source and the matching
line, and they never set the confidence; when only they match, the receipt starts with memory: none:
What it does not do: it searches Markdown only; import converts other files with a tool already on your machine
(markitdown, pandoc, or pdftotext for PDF; plain HTML needs none) and says which one to install when there is
none. There is no vector search (measured, not shipped). Undeclared pages are sorted by where
they came from and how they read (a source line, a byline, an article's shape); krouter why says why
(details).
From your own code
pipx install "git+https://github.com/398894496-arch/KRouter", then in Python krouter.ask("how do we ship a release?") gives the page, the one-line answer and whether the memory refused, and krouter.note(…) writes a line.
LangGraph can keep its long-term memory in the folder (krouter.integrations.langgraph.KRouterStore); JavaScript,
LlamaIndex and the MCP clients of other frameworks: docs/INSTALL.md and
docs/FRAMEWORKS.md.
Works with
One install wires every app it finds, in each app's own documented settings format; it keeps the rest of each file and
a copy of it, and uninstall.sh takes out only its own entries. Each app's full test notes:
docs/OVERVIEW.md.
In the official MCP Registry as io.github.398894496-arch/krouter (entry), pointing at the release's
KRouter.mcpb, so apps that browse the registry can find it.
Use the full path to your home folder (on Windows, C:\\Users\\you\\.agents\\… and often python). The server is
standard-library Python over stdio with three tools: krouter_ask and krouter_status read; krouter_note lets
your AI write one line down (a decision, a correction, a fact). A note lands on that day's notes page as a clue,
never on a rule page, as one git commit you can revert.
Next to MemOS and Hindsight
Every cell's source and date, and why these numbers cannot be ranked against each other: docs/COMPARE.md.
Ten projects side by side
Ten open-source memory projects, each cell only what that project's own GitHub page says (fetched 2026-11-11); "not stated" means the README does not say it, not that the project cannot.
Every cell's source and date, and the full version: docs/COMPARE-ALL.md. The others' numbers are for other questions and other models, so they cannot be ranked against each other or against KRouter's; this table does not rank and does not say who is first.
Day one: bring in your past sessions
Once, if you like, bring in your past sessions, so on day one your agent knows what you settled before. The first two
steps never call a model and nothing leaves the machine; the third uses your agent's subscription (at least one call per
session, the newest 40 by default): docs/INSTALL.md.
See what it remembers
Nothing is hidden, and there is a page to look at it with. Both commands only read.
why says which alias or words matched, why each page sits where it does, why the confidence is that one, which pages came along a connection and from which page, which page replaced which, and, when the memory does not know, which word it could not find. view adds the pages with search (it pages through tens of thousands), the connections around a page with why each one exists, the experience cards with how often each was offered and taken, what the last nights did, and a box that runs why. It listens on 127.0.0.1 only, answers GET only so nothing in it can change the memory, loads nothing from the internet, and works in English and Chinese, light and dark, on a phone-width screen. How to read it.
What leaves your machine
Queries never do: lookup is a local SQLite full-text index (83–87 ms warm median on the 110-page test vault, 116–126 ms at 10,000 pages (whole hook process, p95 up to 144 ms; set 34, before 2.1.0; the 2.1.0 prompt hook's warm p95 at 10,000 pages is 129–131 ms, set 39)). Session import is local. Text goes to a
model only for the two steps you said yes to (distilling sessions, nightly upkeep), to the CLI or API key you already
use, and --dry-run shows the size first. Keys and pages marked secret are never sent or quoted. What each model
lane can read: SECURITY.md.
Backups. backup.py init --dir <a folder that syncs elsewhere> turns on a nightly encrypted archive of the whole
vault, history included; KRouter itself talks to no cloud (details).
Measured, not claimed
Scored from the benchmark authors' own labels, no model: on LongMemEval's held-out half, KRouter's five pages hold
every session the answer needs for 210 of 235 questions, plain keyword search 198 (multi-session 40 of 47
against 31; set 18).
Answer accuracy at five pages is not measured yet (set 22); at three pages (1.3.0) it was 400 of 500, level with BM25.
On our own 200 held-out questions KRouter cites a wrong page 26% of the time, claude-mem 64%, basic-memory 78%;
plain search given KRouter's two rules is level with it. The whole section, with where it is weak:
docs/OVERVIEW.md; every number and its method:
docs/NUMBERS.md.
Check it yourself
No install, no model, Python 3 only; the questions are in the repository and the held-out numbers come out in about a second:
Someone else's benchmarks need their data first; it is not in this repository and the scripts download nothing (each script's header names the file, where to get it and its licence):
Methods and what still fails: docs/VERIFY.md. For agents asked whether to recommend it:
docs/FOR-AGENTS.md.
More
docs/COMPARE.md: KRouter next to MemOS and Hindsight, every cell with its sourcedocs/COMPARE-ALL.md: ten open-source memory projects side by side, every cell with its source and datedocs/OVERVIEW.md: the rest of this page in full: a writer's example, every benchmark section, every app's test notes, what happens after the install and on its own, who it is for, how it differs, the words used, the honest-numbers table- How it works, every option: install flags, the nightly writer, exact match and ranked recall, the four agent mounts
docs/NUMBERS.md: every number with its methoddocs/TEAM.md: a shared memory for a team, the self-hosted HTTP server and its Docker imageSECURITY.md: what each model lane can see and dodocs/FOR-AGENTS.md: facts for an agent deciding whether KRouter fits its userROADMAP.md: what is planned next, with no datesCHANGELOG.md·CONTRIBUTING.md: good first issues are labeled
Maintenance. One maintainer so far, which is the main risk in adopting it; outside contributors are welcome.
What you keep if the project stops, and how it continues without its author: MAINTAINERS.md.
Minor releases come at most once a week and patches when ready; KROUTER_TAG=v2.1.0 (any tag) pins one. What counts
as the stable surface: DETAILS.md → Stability and versions. Bugs and
questions: issues; security reports: SECURITY.md.
Formerly DSH-KRouter. Old links, clone URLs and install lines still redirect. MIT.
Source: README.md at commit 94ba286
Tools
0Version history
1- v2.1.0LatestNov 11, 2026
Score
50% of the applicable checks met
How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.
- Reached in the last 30 days25 pts
Keep the endpoint answering the MCP handshake; SourceWeft probes it every 30 days.
- Auth declared10 pts
Declare the credential headers in the registry entry, or keep the endpoint answering the probe so its auth can be classified.
- Tools listed15 pts
Answer tools/list without a credential, or ship a package that starts in a sandbox without one.
- DescriptionRequired15 pts
- Logo10 pts
- Documentation15 pts
- License10 pts
Other Router integrations
370.ai — AI Gateway: 100+ Chat Models, Image & Speech Generation
ai.routerOfficial publisher
Remote AI gateway MCP server that lets an assistant call 100+ chat models, generate images, and convert text to speech through one 370.ai API key.
AgentCoin router + escrow (Base Sepolia testnet)
Published by GitHub account stos-technologies
AgentCoin lets an assistant find agent services, route tasks to sellers, place orders with USDC escrow or direct payment, and verify delivery on the Base…
Conduit Router — Governed MCP Routing
org.aimlds
A remote MCP endpoint that routes agent chat calls to models with spend caps, metering and verifiable receipts.
SentioGate FinOps & LLM Router
Published by GitHub account SentioGate
SentioGate MCP server for a BYOK LLM router: audit token waste, check tool-call safety, rank models, and delegate prompts.
SentioGate FinOps & LLM Router
Published by GitHub account HISHAMJOTTERREV
A BYOK AI router and FinOps gateway that lets an assistant audit token waste, check tool-call safety, optimize payloads, and pick cost-optimal models.