Every workspace we run is assembled from a bin of parts, not
written for one client. This page is the bin's own account of itself: 11 parts,
5 roles we can be asked for, and — computed, not claimed — which of them the bin can
actually assemble today.
sign in
describe the agent
proposal
operator approval
$5 by x402
your workspace
1 of 5 roles the bin can assemble today. Tell the designer what the job is and it turns
that into a proposal — the role, the parts, the connections and the limits — in about five
exchanges: design an agent.
Answers a customer from the client's own documentation, and escalates what it cannot.
connections/emailnot built the support mailbox the client owns
external/graphifyGraphify (wrapped) — knowledge-graph retrieval over a corpus retrieval over the client's own corpus — the answer must come from their documents, not from the model's memory
documents/extractReading floor (pure Python) reads whatever the customer attached
documents/reportReport renders the summary when a case is escalated to a person
connections/emailnot built the support mailbox the client owns
deepseek.x402.press phrases the answer — paid per call
qwen.x402.pressoptional a second model before it says 'we cannot help with that'
Follows up a lead with the facts, and keeps the pipeline honest about what was actually said.
connections/emailnot built outbound and inbound, from a mailbox the client owns
documents/extractReading floor (pure Python) reads what the prospect sent
documents/reportReport renders a proposal tailored to the prospect
documents/data_tableData table the comparison the prospect asks for
documents/officeDocument engine (pure Python) the letter or deck that goes out
connections/emailnot built outbound and inbound, from a mailbox the client owns
connections/whatsappoptionalnot built where a lot of prospects actually reply
deepseek.x402.press drafts and reasons — paid per call
news.x402.pressoptional what the prospect's market did this week
The other 4 are each waiting on connections/email — one part, not a rebuild.
The bin
documents5
documents/data_table
Data table Render a titled spreadsheet with a header row and as many data rows and columns as the caller supplies.
0.1.0
documents/extract
Reading floor (pure Python) Read a PDF, DOCX, XLSX, HTML, text, markdown or CSV document into its own text — the reading half of the documents class, with scans and images left to the heavy path.
0.1.0
documents/invoice
Invoice Render a sales invoice with seller and buyer details, line items, tax and totals.
0.1.0
documents/office
Document engine (pure Python) Render a backend-neutral template spec plus its fields into PDF, DOCX, XLSX, HTML or plain text: no external process, and byte-identical output for the same input.
0.2.0
documents/report
Report Render a long-form report with a summary block and repeating named sections.
0.1.0
compute3
compute/fundamental_analysisvetted
Fundamental analysis engine Compute risk and return metrics, financial-statement structure, valuation multiples, DCF fair value and factor exposures from caller-supplied data.
0.1.0
compute/macro_analysisvetted
Macro analysis engine Compute yield-curve and spread structure, breakeven inflation, growth rates, inflation decomposition, FX conversion, rates maths and real-commodity prices from caller-supplied data.
0.1.0
compute/technical_analysisvetted
Technical analysis engine Compute technical indicators, volume and order-book metrics, chart patterns, ratios, signals and backtests from caller-supplied OHLC data.
0.1.0
external2
external/graphify
Graphify (wrapped) — knowledge-graph retrieval over a corpus Turn a folder of documents into a queryable knowledge graph, then answer questions by traversing it — retrieval for a corpus that has no vector store.
0.1.0
external/wolfram_alphaclient
Wolfram|Alpha (wrapped) — curated knowledge, Wolfram Language evaluation, chart artefacts Answer a factual question across the ~40 curated domains our own fleet owns nothing in, evaluate Wolfram Language to an exact value, and get a chart back as a PNG — every result carrying its interpretation, the link to that specific result page and the attribution Wolfram's terms require.
0.1.0
connections1
connections/telegramclient
Telegram Hold a client's own Telegram bot open, so their agent can be reached in the chat they already use — and only from the chats they paired.
0.1.0
The worked example
One role with one client's specifics — a tenant manifest. This is what a
proposal has to produce, and the only one we have assembled end to end.
Deep Research deep_researchclient data
Research an open question across the web and the x402 service fleet, then deliver a cited long-form report as a document.
plain, factual, citation-first — no filler and no confident guesswork Anything needing a decision, a payment above the profile budget, or contact with anyone outside the studio goes to the studio owner in the portal.
Ready to describe the job?
Sign in with an email code and tell the designer what you need. About five
exchanges, then a proposal you can approve — $5 in USDC over x402, credited to the workspace.