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

# Tool layer and agent access

Extend the hands, not fork the brain. The same stores, for any agent.

Everything the desk reads is also available as **tools**: read-only, point-in-time endpoints over the engine's stores. The desk's own agents use them, and so can outside agents and integrations.

## The tools

| Tool               | Returns                                                          |
| ------------------ | ---------------------------------------------------------------- |
| Market state       | The anchor, regimes, fired flags and structural signatures.      |
| Token factor state | One asset's percentiles across windows and its regimes.          |
| Factor series      | The recent observations of one stream, for charting or checking. |
| Divergences        | Fired flags, filterable by kind and asset.                       |
| Mechanism          | The mechanism graph as of a given time.                          |
| Corpus search      | Captured articles, with source, time and content hash.           |
| Claims             | Typed claims with claimant and role.                             |

Every tool takes an "as of" time, and nothing returned can postdate it.

## Access today

* **JSON-RPC** over HTTP and stdio, following the shape of the Model Context Protocol.
* **API keys** for integrations. Keys are stored only as hashes and reach the read and ask surface, never account management.
* **The ask endpoint**, as JSON or as a live event stream of stages, evidence blocks and the verified answer.

## Built on SERA

LensAI's base agent is **SERA-CryptoAgent** from Sentient, open source under the Apache-2.0 licence. LensAI extends its hands rather than forking its brain: its stores are wrapped as additional tools that SERA's router can select, and generated drop-in tool modules let SERA call the engine directly. SERA carries none of LensAI's guardrails; those are enforced in LensAI's own narration layer and verified on output. Attribution and notices are preserved as the licence requires.

## Next: a free, standard MCP server

Agents increasingly act on market data without checking it. A prompt hidden in a social post has already been used to make an AI-controlled wallet send funds, and data-API builders report agents acting on numeric fields with no sanity check. LensAI's answer is **verify before act**:

* a free, standard remote MCP server, listed in the official MCP registry and offered to agent platforms;
* Claim Check as a tool, so an agent can check a number before acting on it;
* instructions stripped from any quoted post before it is served;
* published uptime.

No local package will ever ask for a private key.


---

# 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 by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.scanbylens.xyz/the-engine/tool-layer.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

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.