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

# Praxa Execution Fabric

> Stripe routes payments. Praxa routes work — and can prove what it did.

**Stripe routes payments. Praxa routes work — and can prove what it did.**

Praxa Execution Fabric is an API for delegating real work to an AI agent and getting proof of what happened. You submit a task. Praxa routes it across models, tools, and agents under a policy you set. Anything risky (a purchase, a message sent on your behalf, a destructive change) pauses for your approval before it executes. When the run finishes, you get an outcome and a receipt: what ran, which policy version evaluated it, who approved what, and what it was verified against.

It's the same execution runtime that already runs Praxa's own custom agents and its desktop coding tool, opened up as an API you can build on.

Model routing is table stakes now. Every major provider ships a router. What we bill you for is the part that's hard to fake: a policy that actually gates the risky step, and a receipt that proves it did.

## How a task moves through the fabric

```
POST /v1/execute                submit a task, under a policy
        |
        v
   run starts                   routed across models and tools under that policy
        |
        v   (only if the policy calls for it)
   approval.required             event carries the exact action plus an approval digest
        |
        v
   you decide                   POST /v1/runs/:id/approve  { action_digest, decision }
        |
        v
   run finishes                 outcome + receipt: what ran, which policy, who approved, verified how
```

That's the whole shape. Request fields, event types, and the error format live in the API reference below.

## Status legend

Every capability in these docs carries one of three labels. They tell you what you can build on now versus what to plan around.

| Status                | Meaning                                                                                                                                                                                  |
| --------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| **Available today**   | Shipped and callable now. Build on it in production.                                                                                                                                     |
| **Developer preview** | The wire contracts are stable and published, but the serving endpoints are still in active build and not yet publicly callable. Interfaces can still change before general availability. |
| **Planned**           | Designed and sequenced, not yet built. See [`status-and-roadmap.md`](/fabric/status-and-roadmap) for what has to happen first: timing is phase-gated, not date-committed.                |

## Documentation

### Getting started

* [`getting-started/overview.md`](/fabric/getting-started/overview) — what the fabric is and how a request flows end to end
* [`getting-started/concepts.md`](/fabric/getting-started/concepts) — runs, events, receipts, policy, and tenancy in one page
* [`getting-started/quickstart.md`](/fabric/getting-started/quickstart) — make your first authenticated call

### API reference

* [`api/execute.md`](/fabric/api/execute) — `POST /v1/execute`: submit a task, choose turn or task mode
* [`api/runs.md`](/fabric/api/runs) — `GET /v1/runs/:id`: read a run's current state
* [`api/events.md`](/fabric/api/events) — `GET /v1/runs/:id/events`: stream a run's events, resume with `Last-Event-ID`
* [`api/approvals.md`](/fabric/api/approvals) — `POST /v1/runs/:id/approve`: resume a paused run
* [`api/usage.md`](/fabric/api/usage) — `GET /v1/usage`: tenant-scoped usage and spend readback
* [`api/webhooks.md`](/fabric/api/webhooks) — `POST /v1/webhooks`: register a delivery endpoint
* [`api/errors.md`](/fabric/api/errors) — the RFC 9457 error format every endpoint shares
* [`api/openapi.yaml`](/fabric/api/openapi.yaml) — machine-readable OpenAPI 3.1 specification

### Keys & access

* [`keys-and-access/api-keys.md`](/fabric/keys-and-access/api-keys) — creating, scoping, and rotating `praxa_sk_` keys
* [`keys-and-access/scopes-and-tenancy.md`](/fabric/keys-and-access/scopes-and-tenancy) — tenants, scopes, and the three principal types
* [`keys-and-access/rate-limits.md`](/fabric/keys-and-access/rate-limits) — per-key limits and response headers

### Approvals & policy

* [`approvals-and-policy/approvals.md`](/fabric/approvals-and-policy/approvals) — how a run pauses and how you resume it
* [`approvals-and-policy/policy-engine.md`](/fabric/approvals-and-policy/policy-engine) — the `policy` parameter and how risk tiers get decided
* [`approvals-and-policy/budgets-and-limits.md`](/fabric/approvals-and-policy/budgets-and-limits) — budget ceilings and what happens when a run hits one

### Security

* [`security/security-model.md`](/fabric/security/security-model) — tenant isolation and credential handling
* [`security/data-handling.md`](/fabric/security/data-handling) — what's stored, for how long, and whether any of it trains a model
* [`security/receipts-and-audit.md`](/fabric/security/receipts-and-audit) — receipts and the hash-chained evidence trail behind them

### SDK

* [`sdk/typescript.md`](/fabric/sdk/typescript) — the TypeScript client
* [`sdk/streaming.md`](/fabric/sdk/streaming) — consuming a run's event stream in code
* [`sdk/errors-and-retries.md`](/fabric/sdk/errors-and-retries) — client-side retry and error-handling semantics

### Use cases

* [`use-cases/overview.md`](/fabric/use-cases/overview) — the three markets the fabric is built for
* [`use-cases/agent-builders.md`](/fabric/use-cases/agent-builders) — upgrading an existing custom agent onto the fabric
* [`use-cases/coding-and-desktop.md`](/fabric/use-cases/coding-and-desktop) — cloud execution lanes for desktop coding sessions
* [`use-cases/business-workflows.md`](/fabric/use-cases/business-workflows) — internal org workflows on the same substrate

### Enterprise

* [`enterprise/teams-and-organizations.md`](/fabric/enterprise/teams-and-organizations) — org-level tenancy and billing
* [`enterprise/billing-and-plans.md`](/fabric/enterprise/billing-and-plans) — tiers, metering, and the plan ladder
* [`enterprise/deployment-and-isolation.md`](/fabric/enterprise/deployment-and-isolation) — dedicated isolation, and why there's no self-hosted option

### How-to guides

* [`how-to/build-an-approval-ui.md`](/fabric/how-to/build-an-approval-ui) — render an approval payload correctly (render == record == what actually runs)
* [`how-to/receive-webhooks.md`](/fabric/how-to/receive-webhooks) — verify signatures and handle retries
* [`how-to/bring-your-own-keys.md`](/fabric/how-to/bring-your-own-keys) — connect your own provider credentials
* [`how-to/migrate-from-provider-apis.md`](/fabric/how-to/migrate-from-provider-apis) — moving from a direct model API to the fabric

### Reference

* [`faq.md`](/fabric/faq) — common questions, answered honestly
* [`status-and-roadmap.md`](/fabric/status-and-roadmap) — what's shipped and what's next, in customer terms
* [`glossary.md`](/fabric/glossary) — terms used across these docs

## Support

Questions, feedback, or partner-preview access: *(support contact placeholder — add before publish, e.g. a support address or partner-intake form)*.