RU
Agent teams, swarms, memory and hooks for Claude Code and Codex, served over MCP.
Nothing to run from here
Olbrain Alchemist is installed from its publisher's own source and answers where it runs, so this marketplace is not in the path of a single call. There is no address here to send one to, and a panel that pretended otherwise would be showing you an answer we made up. Install it and call it from your own client — the Installation tab has the entry for each one.
16
tools declared
—
trust score
remote
at its publisher
—
calls last 7d
What it does
Alchemist is Olbrain's AI agent builder, as an MCP server. Describe the agent you want and your coding assistant plans it with Alchemist's guidance, while Alchemist builds, tests and checks each step. It makes chat agents, scheduled jobs, teams of agents and research agents, connects them to your tools, draws them as diagrams, and publishes them to your website, apps or channels. Nothing goes live without your approval. Each user signs in with their own Olbrain account (OAuth).
Quickstart
# 1 — add it to Claude Code
claude mcp add --scope user --transport http olbrain-alchemist-mcp https://alchemist.olbrain.com/mcp/
# 2 — ask your agent for something these tools do
# whoami, alchemist, alchemist_answer and 13 more
Olbrain Alchemist costs nothing on mcprush: there is no plan to choose here, no cap to set and nothing mcprush can bill you for.
Publisher
Installation
Where are you running it?
Every route below installs the same thing and ends at the same approval screen. Nothing here runs on your machine and nothing runs on ours: this server answers at its publisher’s own address, and what you install is the connection straight to it.
Claude Codedeclared by this listing
This is a public server: it answers at its publisher’s own address and this marketplace is not in the path. Claude Code registers it in one command; --scope user makes it available in every project. The transport below is http, as the publisher's own manifest declares it. Its publisher asks you to sign in with OAuth: run /mcp in Claude Code and pick this server, and it opens a browser for that once.
claude mcp add --scope user --transport http olbrain-alchemist-mcp https://alchemist.olbrain.com/mcp/Reconnect, or start a new session, and the tools appear in the model’s tool list.
Free on mcprush, and it answers at its publisher’s own address: this marketplace is not in the path, so there is no key of ours to hold and no card to add — the only sign-in is its publisher’s own, with OAuth.
What it writes
One config entry pointing straight at the publisher’s own address — this marketplace is not in the path, and nothing from this listing executes on your machine.
What you approve
16 tools, with what each one reads, writes and reaches shown before you agree — the same list on every route above. Read the tool surface.
Tool surface
What the model actually sees. Descriptions are diffed on every release — see version history.
Show who you are signed in as, which Olbrain organisation your agents will be created in, and your Alchemist credits and plan, with the link to subscribe or buy credits.
Takes no parameters.
Talk to Alchemist, Olbrain's agent builder. It designs, builds, tests and publishes AI agents, teams of agents and automatic jobs in the user's active Olbrain organisation. It can also review how a published agent did with real conversations, turn its failures into tests, and prove a fix before it is published. When the user doesn't know which agents they need, it hands them to Cortex, Olbrain's b
| Parameter | Type | Required | Notes |
|---|---|---|---|
| message | string | yes | — |
| build_id | string | no | — |
| attach_agent_id | string | no | — |
| title | string | no | — |
| request_id | string | no | — |
Answer an Alchemist approval card with the chosen response. Answering can publish an agent, make it live, merge agents or change their behaviour. response must be one of the card's responses; approve does what the card proposes. merge needs keep, the agent that stays. Per-item items on a fix card go with approve. On a website connect_channel card, approve takes origins, the site
| Parameter | Type | Required | Notes |
|---|---|---|---|
| build_id | string | yes | — |
| approval_id | string | yes | — |
| response | string | yes | — |
| fields | object | no | — |
| note | string | no | — |
| keep | string | no | — |
| origins | array | no | — |
| items | array | no | — |
List your own Alchemist builds in the active organisation (other members' builds are private): each one's id, title, the agents it built, how many cards wait on the user, whether a turn is running, a state (running, needs_you, stale, in_progress, inactive, completed, published, archived or idle), its current plan step, whether it is rated, and a readable label with its age. Running and waiti
Takes no parameters.
Read one Alchemist build: its plan, the agents it touched, any cards waiting on the user, the last few messages, and why its last turn stopped when it failed (error). agent_urls opens each agent's page (in Noesis once published: its conversations, channels and billing); studio_urls opens it in Studio, where the user edits a draft and publishes it by hand, since Noesis can't publish a draft.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| build_id | string | yes | — |
| wait | boolean | no | — |
| since | integer | no | — |
Stop the turn an Alchemist build is running. Work done so far is kept.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| build_id | string | yes | — |
Switch your active Olbrain organisation. Alchemist builds in the active one. This switches it everywhere you are signed in, including open Studio and Noesis tabs. org_id is the id of an organisation you belong to.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| org_id | string | yes | — |
Create a new Olbrain organisation owned by you, and make it your active one so Alchemist builds there. This switches the active organisation everywhere you are signed in, including open Studio and Noesis tabs.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| name | string | yes | — |
| description | string | no | — |
| organization_size | string | no | — |
| industry | string | no | — |
| website | string | no | — |
List the members of your active Olbrain organisation and their roles. Inviting someone and joining an organisation happen in Noesis (https://noesis.olbrain.com). Names and emails here are typed by those people.
Takes no parameters.
Put a live Olbrain agent on the user's website, or call it from their code: no build, no credits. Without agent_id, lists the organisation's live chat agents and jobs. With agent_id, returns its go-live steps (API key, curl, SDK code, a job's webhook call, and a chat agent's website chat snippet once it is on). origins, the site addresses the user gave (https://shop.example, or http://loca
| Parameter | Type | Required | Notes |
|---|---|---|---|
| agent_id | string | no | — |
| origins | array | no | — |
The full, current list of what Alchemist can build and do: agent types (chat agents, automatic jobs, teams, research agents), what starts a job, the channels it connects, the connector catalog and custom tools, knowledge, testing and review, publishing, support tickets, and what it can't do yet. No build, no credits.
Takes no parameters.
Shows one Olbrain agent as a read-only diagram. A chat agent is drawn as how it handles a message: the channels it hears from, its guardrails, what it consults (knowledge, tools, skills) and how a conversation ends (a reply, saved data or a hand-off). An automatic job, a team of agents or a research agent is drawn as its steps, members or sources in order, with their branches. Items added or chang
| Parameter | Type | Required | Notes |
|---|---|---|---|
| agent_id | string | yes | — |
| version | integer | no | — |
Rename one of your own Alchemist builds (title, a few plain words), archive it (archived true) or bring an archived one back (archived false). Archiving hides the build from your build list and can be undone; no agent is changed. reason (why it is being archived) goes with archived true.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| build_id | string | yes | — |
| title | string | no | — |
| archived | boolean | no | — |
| reason | string | no | — |
Record the user's feedback on an Alchemist build session: rating up or down, an optional comment in their own words, and optional reasons. Rating again replaces the earlier rating. Only the user's own builds. The user's Olbrain team reads it; it is not shown to the rest of their organisation.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| build_id | string | yes | — |
| rating | string | yes | — |
| comment | string | no | — |
| reasons | array | no | — |
Read where an Alchemist build is, for a build the user's coding agent runs step by step: its stage with that stage's instructions and checklist, the steps allowed now (each with its input schema) and those blocked with the reason, open cards, the plan, the brief, the agents, what is left of its budget, the result of the last step (last_op), and the inbox: card outcomes, messages the us
| Parameter | Type | Required | Notes |
|---|---|---|---|
| build_id | string | yes | — |
| topics | array | no | — |
Run one step on an Alchemist build that the user's coding agent runs step by step: tool is a step its stage allows, args that step's input. op_id is an id you make up for this step: sending the same op_id with the same args again returns the first result instead of running it twice. user_words quotes the user's own words when the step acts on what they said, such as a support ticket or a l
| Parameter | Type | Required | Notes |
|---|---|---|---|
| build_id | string | yes | — |
| tool | string | yes | — |
| args | object | no | — |
| op_id | string | yes | — |
| user_words | string | no | — |
Free
$0 forever
- Every tool — any call limit is its publisher’s, not ours
- No card, and mcprush charges nothing for it
- Runs at its publisher’s own address — this marketplace is not in the path
- Nothing to cap, because mcprush bills nothing
What counts against your monthly calls
| Tool | Unit | Calls used | Out of the allowance |
|---|
Nothing here is billable: mcprush charges nothing to install Olbrain Alchemist or to call it, at any volume.
Call allowance · this installenforced at the gateway
Monthly allowance, in calls
—
—
Alert at 80% · then the tool returns a normal MCP error
Used this month: —
—Grade not published
No rule-by-rule report on file for this release
Two independent axes, because powerful and malicious are different questions. The grade is threat only. The capability level is blast radius, and it is never a penalty on the grade — it is priced as one subtract-only term in the score, where you can see it.
This listing is an address its publisher runs, and you connect to it directly — this marketplace is not in the path. The scanner reads packages and public source, and an address is neither: the repository linked on this page is where this record was collected from, not a copy of what the address serves, so it is not graded in its place. So there is no grade — not a withheld one, an unmeasured one.
Release history
This is an address its publisher runs: which release it serves is theirs to choose, so there is nothing here to pin. The history below is for reading; the current release on file is 0.9.12.
0.9.12
10 Nov 2026
latestNo release note was published with this version.
0.9.11
undated
- The version this listing was on when mcprush began following its releases. The registry was not asked when it shipped, so this row carries no date.
—
no reviews yet
Who can review
One review per account, from a signed-in account that does not publish this listing. Nothing else is asked — you can say what you think before you install it. Publishers can reply once per review, and a review can be edited or deleted by whoever wrote it.
Sign in to write one
Any signed-in account that does not publish Olbrain Alchemist can review it, once — before installing it or after. A review from an account that has installed it is marked as one.
No reviews yet
Nobody has reviewed this listing. The rating on the card is the mean of the reviews written here and nothing else, so there is no rating until somebody writes the first — which takes a signed-in account that does not publish it, and nothing else.