> ## Documentation Index
> Fetch the complete documentation index at: https://docs.altss.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Intelligence Workspaces

> Plan entity tables, research columns, evidence and budget for individual or managed batch work.

An Intelligence Workspace organizes an agreed research universe and its outputs into a reviewable table. It connects the decision, entity identities, selected operations, source evidence and budget. Annual App workspaces and Console metered execution retain their separate entitlements.

## Define the table

Start with companies, investors, people or funds that you are authorized to provide. Give each row a stable identity, entity type and the identifiers needed by the selected operation. Do not assume that similar names refer to the same legal entity. Keep customer-provided notes separate from independently sourced facts.

A useful initial schema is:

```json theme={null}
{
  "columns": [
    "entity_id",
    "entity_type",
    "name",
    "canonical_reference",
    "research_question",
    "operation_id",
    "input",
    "quote_status",
    "maximum_budget",
    "run_id",
    "result",
    "source_urls",
    "observed_at",
    "verified_at",
    "unknowns",
    "review_status",
    "final_charge"
  ]
}
```

This is a planning schema for your own table, not a newly introduced Console endpoint or a claim that every column has an automatic executor.

## Add research and enrichment columns

Select operations from the [source-backed catalog](https://docs.altss.com/console/capabilities.md). An available operation can be used only with its actual inputs and permitted output fields. Coming-soon columns cannot be run yet. Keep each result's evidence, dates and limitations attached to its row.

## Price the workspace

[Compare the agent and table tools](https://docs.altss.com/console/prices.md#agent-and-table-tools). Intelligence tables use underlying operation prices; each table cell is charged the applicable successful-operation price of its capability. Do not charge both a package and its components by assumption. Custom workflows and enterprise monitoring are custom scope and require a proposal.

## Batch execution and budget

For `company.financial_research`, quote and accept each eligible issuer separately with a distinct idempotency key for each accepted attempt. Sum proposed maximum charges before approval and stop when the remaining budget is insufficient. Agree other batch scope and budget before work starts.

## Conditional execution

Define conditions such as identity resolved, mandatory input present, quote within budget and prior result reviewed. Use supported client-side orchestration for the current per-issuer contract. The public contract does not establish a hosted general table executor or atomic batch transaction; do not describe a planned condition as a completed run.

## Save and monitor a workflow

Preserve the universe definition, input version, operation schema/version, accepted quotes, idempotency keys and review decisions in your own authorized workflow. For recurring monitoring, agree the event types, cadence, source coverage and maximum budget. A saved brief does not itself schedule execution or guarantee source-complete alerts.

## Review and export

Separate facts, derived analysis and unresolved questions. Keep observed dates, source publication dates and verification dates distinct. Export only customer-owned inputs and delivered outputs permitted by their terms. This workflow does not grant bulk export of the Altss database or replication of its relationship graph.

[Define a managed workspace](https://altss.com/subscribe?product=agents-scoped-task\&useCase=RESEARCH\&source=%2Fworkspaces) · [Compare workspace delivery](https://altss.com/workspaces)


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.