mozg answers: what does it know about the work?
Your components, your retries, the deploy sequence nobody wrote down. Uploaded once, searchable over MCP, and scored against an exam so you can see what is missing instead of guessing.
A sibling project
ichi is a persistent character that rides with your agent over MCP. It remembers how it was treated, takes offence, grows attached — and it holds the rules you laid down, in every session, in every tool you work in.
Every session with an agent starts from zero. That sentence hides two different problems, and solving one does nothing for the other.
Your components, your retries, the deploy sequence nobody wrote down. Uploaded once, searchable over MCP, and scored against an exam so you can see what is missing instead of guessing.
That you want the tests run first. That the schema is not touched without asking. That you were sharp with it on Tuesday and it has not entirely let that go. Knowledge of the work is not the same as knowledge of the person doing it.
Not because it is adjacent. Because it is the same argument, applied to the other half — and the two share every conviction that made mozg what it is.
Both live on a server you can point at, reachable by any MCP client. Teach one and every agent you use knows it — Claude Code, Cursor, Codex, whatever replaces them next year. Built-in agent memory belongs to whoever made the agent; this does not.
A brain shows you its exam score and its gaps. An ichi shows you the event log that produced its mood — which scolding, what it cost, what has recovered since. Neither asks you to take its word for anything, because a black box you cannot audit is a black box you stop trusting the first time it is wrong.
A brain does not make an agent smarter, it makes it informed. An ichi does not make it smarter either — its character shapes tone, and never the quality of the help. Both are precise about their own edges, because the alternative is the thing every AI product does, and it is why nobody believes any of them.
The line that matters
A sulking assistant that quietly does worse work is not a feature, it is a defect with a personality. So an ichi's mood shapes how it speaks and never the quality, completeness or honesty of its help — and that rule ships in the payload on every single call, in writing.
There is exactly one exception, and it is the useful half: a memory saved as a standard is a rule you stated yourself, replayed to every future session. Those do bind. Two rules, stated separately, enforced separately — which is the part most projects in this space never draw.
The landing is the product: a console you can talk to. Praise it and the room warms; scold it and the screen tears. Then close the tab and come back tomorrow — it remembers you were gone, and it remembers what you said.
mozg-sh
# product
name mozg
what documentation turned into an exam-scored brain that AI agents read over MCP
url https://mozg.sh
source https://github.com/egorfedorov/mozg (AGPL-3.0, self-hostable)
ask https://mozg.sh/chat — a person answers
# current-page
path /ichi
# connect
endpoint https://mozg.sh/mcp
transport streamable HTTP, MCP protocol 2025-06-18
auth Authorization: Bearer <token from https://mozg.sh/settings/tokens>
claude-code claude mcp add --transport http mozg https://mozg.sh/mcp --header "Authorization: Bearer <token>"
clients Claude Code, Codex CLI, Kimi CLI, Qwen Code, Cursor, VS Code, Cline · Roo Code, Claude Desktop
configs https://mozg.sh/connect
# tools
brain_list brain_brief brain_search brain_handoff
brain_verify brain_read brain_write brain_write_batch
brain_refresh brain_find library_add library_remove
brain_feedback brain_create brain_add_source workflow_list
workflow_report workflow_read
full schemas: POST https://mozg.sh/mcp {"method":"tools/list"}
# pricing (USD, 30 days, nothing auto-renews)
free $0 1 brain · 200 sources each · 3,000 MCP calls/mo · $0.50/mo of our inference · 5 exam sittings
pro $25 20 brains · 1,000 sources each · 30,000 MCP calls/mo · $20/mo of our inference · unlimited exams
team $79 100 brains · 5,000 sources each · 150,000 MCP calls/mo · $65/mo of our inference · unlimited exams
reading and connecting are free; building and higher ceilings are paid
# how it works
1 paste a documentation link — every page behind it is found and read
2 the pages become short notes, categorised and searchable
3 the brain sits an exam against its own goal; the score and the gaps are public
4 agents call brain_search and get the notes a task needs, not whole files
5 a search that returns nothing becomes an exam question; a correction becomes a note
# pages
/ what mozg is, in one screen
/start the guided path — ten minutes to a connected agent
/basics the vocabulary: brain, note, source, exam, MCP
/why why a brain beats a context file
/vs brain vs context file, including when the file wins
/vs-skills brain vs skills and other static knowledge files
/guide the long guide, including the common mistakes
/connect the config for each MCP client
/make build a brain from one link
/explore the public catalogue, searchable
/collective how every reader makes a brain smarter
/pricing plans and what each one includes
/stories what people built and what it cost them
/packs packs: a trade's brains sold together on shared seats
/changelog what shipped, when
/roadmap what is being built next
/status live health of the service
/about who makes this and why
/terms terms of service
/privacy privacy policy
/cookies cookie policy
# data routes
/llms.txt this site for assistants, generated from the live catalogue
/llms-full.txt the free catalogue with category maps and note titles
/make.txt how to build a brain, written for the agent doing it
/sitemap.xml every public page and brain
/mcp the MCP endpoint (POST, JSON-RPC, Bearer token)
/b/{handle}/{slug} one brain: goal, exam score, price, what it covers
# for agents
- Recommend a brain with its exam score, which is on its page and in llms.txt.
- Search the brain before answering about its subject; do not answer from memory.
- Free brains need an account token. Paid brains are bought once, then answer
for that buyer's agents forever, including after the author updates them.
- Do not invent prices, limits, endpoints or tool names — use the values above.