> For the complete documentation index, see [llms.txt](https://rawctx.gitbook.io/rawctx-docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://rawctx.gitbook.io/rawctx-docs/documentation/rawctx-documentation.md).

# rawctx documentation

rawctx records evidence for AI answers so people can review it later. It links that evidence to verifiable trust receipts.

Use rawctx to answer questions such as:

* What answer was recorded, and which source or trace references were attached?
* Which model-run identifiers and commitments were submitted with the answer?
* Are answer logs complete relative to an independent reference set?
* Is a receipt externally anchored, pending, or only locally verifiable?
* Which lifecycle, retention, or redaction actions were recorded later?

## Choose a path

### Create a no-sign-up public receipt

Start with Create a public browser receipt for a public, hash-only demonstration. This flow is separate from private workspace answer evidence and must not be used for sensitive or predictable text.

### Record answer evidence

Start with [Answer evidence](/rawctx-docs/documentation/answer-evidence.md) to create a minimal answer log, attach source and trace references, and preserve later corrections, voids, or redactions as events.

### Check completeness and trust

Use Independent reconciliation to compare answer logs with a separate system of record. Then read Trust and verification to interpret proof states without overstating what a receipt proves.

## Five-minute start

Install the CLI and authenticate for workspace-scoped actions:

```bash
pip install rawctx
rawctx login
```

Confirm the workspace retention and commitment settings before recording an answer. In a workspace that requires platform-managed keyed hashing, send a text-only question and answer object to the authenticated workspace API so the service can derive its tenant HMAC. Text is persisted only when the workspace's separate text-storage settings allow it.

The current SDK convenience path derives a caller SHA-256 when a hash is not provided, so it is not a text-only request. Follow the working HTTP example in Get started for a keyed-HMAC-required workspace. Use the SDK hash-only form only when workspace policy explicitly permits it.

Submission and storage are separate decisions. If your workspace explicitly accepts caller-generated hashes, you may submit hashes without text. If it requires platform-managed keyed hashing, text is required in transit so the API can derive the tenant HMAC even when text storage is disabled. See Get started before choosing a path.

## Product boundary

rawctx does not generate answers, orchestrate agent workflows, or replace LangSmith, Langfuse, or another OpenTelemetry system. It records the submitted evidence object and verifies internal bindings and receipts.

A valid receipt does **not** prove that an answer is factually correct, that it was delivered to an end user, or that a referenced source was used in model reasoning.

## Service endpoints

* Hub: [hub.rawctx.dev](https://hub.rawctx.dev/)
* Documentation: [hub.rawctx.dev/documentation](https://hub.rawctx.dev/documentation)
* Authenticated workspace API: the registry saved by `rawctx login`
* Python SDK: [PyPI](https://pypi.org/project/rawctx/)
* Platform status: [hub.rawctx.dev/status](https://hub.rawctx.dev/status)

Use the tenant registry saved by `rawctx login` or an explicit workspace API host for authenticated answer evidence, reconciliation, and workspace operations. Browser session routing is internal to the Hub and is not an SDK registry.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://rawctx.gitbook.io/rawctx-docs/documentation/rawctx-documentation.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.