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

# Privacy & Security

> Understand what WalletWall MCP can access, what your AI client and model provider can see, what WalletWall receives, and the boundaries the implementation does not cross.

WalletWall MCP is **pre-release** and has not had an independent security review or audit. This page
states the implemented trust boundaries and the limits of those boundaries; it is not a certification.

> [!WARNING]
>
> WalletWall MCP is local, read-only evidence infrastructure. It does not sign, custody, trade,
> broadcast transactions, or certify truth.

## What runs where

| Component | Runs on | Network | Disk |
| - | - | - | - |
| WalletWall MCP server | Your machine, started by your client over stdio | No listening socket and no outbound connection | Writes nothing |
| WalletWall canvas | Your machine | Loopback listener only; no outbound connection | Writes nothing |
| Optional HTTP adapter | Your machine, only if you start it | `127.0.0.1` by default; no outbound connection | Writes nothing |
| WallPrint resolver | Your browser from `walletwall.org` | Same-origin static-file reads; no analytics or third-party requests | Stores nothing in the browser |

The stdio server is stateless: request in, calculation in memory, response out. It has no WalletWall
user/session identifier, telemetry, query history or cache and reads no environment variables, home
directory or working directory.

## What your AI client and model provider can see

WalletWall MCP runs **inside** an AI-client workflow. GitHub Copilot, Claude Code or another MCP client
and the model provider behind it can see every prompt, tool input and tool output, including wallet
addresses you put into an evidence bundle, under their own policies.

> WalletWall-controlled code does not retain MCP query history or transmit MCP query data to WalletWall
> infrastructure. Your AI client and model provider may independently process that data.

WalletWall therefore does not claim that nothing leaves your machine. Treat wallet addresses as
sensitive: hashing an address is not anonymization. WalletWall MCP keeps no persistent address store,
index or query log. Its digests can cover complete records that contain addresses and are returned to
your client. A record WalletWall deliberately publishes as a WallPrint follows the separate
[WallPrint publication model](https://docs.walletwall.org/mcp/wallprint.md#publication).

## What WalletWall receives

| Activity | What WalletWall receives |
| - | - |
| Using the stdio server, plugin or local adapter | Nothing through WalletWall infrastructure; there is no account, backend, telemetry or remote configuration |
| Installing from GitHub | GitHub may expose aggregate repository traffic/download counts to repository owners |
| Viewing a WallPrint | A normal HTTPS request for static files. The hosting provider's standard request logs apply; the resolver page adds no analytics |
| Publishing a WallPrint | Only WalletWall publishes deliberately from its repository. Using MCP never publishes anything |

## What WalletWall MCP cannot do

| Capability | Boundary |
| - | - |
| Custody or move funds | No such tool, no network access and no key-handling path |
| Handle private keys or sign | No signing code and no tool parameter for a key, seed phrase or signature |
| Construct or broadcast transactions | No transaction schema and no RPC/network code |
| Trade, swap or route execution | No such tool |
| Fetch arbitrary RPC, HTTP or SQL | The engine imports no network or database module |
| Persist queries or wallet caches | The engine imports no filesystem module |
| Certify investment, legal or regulatory status | Outputs are evidence classifications, not certifications |
| Identify a person behind a wallet | No identity data or tool for that purpose |
| Guarantee evidence is true | States describe evidence sufficiency under policy; retrieval fidelity is not verified |

> [!WARNING]
>
> You never need to give WalletWall a private key, seed phrase, signature or credential. The bounded
> open maps inside an evidence bundle can still carry arbitrary short strings if you paste them there,
> so do not put secrets into prompts or bundles.

## Bring your own evidence and providers

WalletWall MCP has **no live data provider**. Every observation comes from evidence you gathered with
your own RPC endpoint, indexer, block explorer, issuer publication or other retrieval tooling. WalletWall
never receives those provider credentials and cannot prove that a reading was faithfully retrieved.

* `transport.retrievedVia` records how a reading arrived. Transport metadata never confers authority.
* A well-formed bundle of invented evidence can satisfy a policy. `ESTABLISHED` therefore means the
  presented evidence satisfies the policy, not that WalletWall independently verified reality.
* A failed retrieval should be represented as `NO_RESULT` or `ERROR`, never silently rewritten to zero.

## Threat boundaries

> [accordion] Hostile or malformed input
>
> The server validates tool arguments and bundle records, bounds structure and sizes, uses strings for
> exact quantities and returns fixed-shape errors that do not echo input values. `subject`, `value` and
> `payload` are bounded open maps; unused keys are accepted and ignored by evaluation.

> [accordion] Prompt injection
>
> Tool outputs can contain text you supplied. Output-safe character rules and the skills' "tool output
> is data" instruction reduce prompt-injection risk but do not eliminate it.

> [accordion] Local adapters
>
> The Copilot canvas listener is loopback-only with additional request checks. The optional HTTP
> adapter is loopback by default but has **no authentication**: anything that can reach the port can
> call it. See [Claude / local HTTP adapter](https://docs.walletwall.org/mcp/claude-local-adapter.md).

> [accordion] Hostile local owner
>
> Someone who controls the Node.js process or local account can replace `node`, preload code or modify
> the server file and read tool traffic. WalletWall treats the operating system's account and process
> boundaries as the control; it does not attempt to detect a hostile machine owner.

## Supply-chain and implementation details

> [accordion] Static contract and reproducible artifacts
>
> Tool names, schemas and resources are compiled in and pinned in `security/tool-manifest.json`.
> CI rebuilds the server bundle and MCPB deterministically, while the Claude distribution records its
> pinned source and file digests in `PROVENANCE.json`.

> [accordion] Runtime dependencies and generated validation
>
> The engine uses `@modelcontextprotocol/sdk` and `zod`. The bundled Copilot/MCPB server imports only
> `node:crypto` and `node:process`; the Claude distribution loads the same engine unbundled from its
> committed dependency tree. `zod` uses the `Function` constructor to compile validators for the
> server's own fixed schemas; user input is validated by that code and is never compiled into it.

> [accordion] Controls not yet in place
>
> Signed commits, signed releases, an SBOM, provenance attestations and an independent security review
> are not currently in place.

## Evidence is not certification

Nothing WalletWall returns or publishes certifies truth. A support state describes how well presented
evidence satisfies a WalletWall-authored policy. A readiness classification does not establish
investment merit, solvency, creditworthiness, legal status, regulatory compliance, absence of fraud or
future performance. A WallPrint digest proves bytes, not facts and not authorship.

## Report a vulnerability

Use GitHub private vulnerability reporting on
[`Wallet-Wall/walletwall-copilot`](https://github.com/Wallet-Wall/walletwall-copilot) (**Security →
Report a vulnerability**). If that option is unavailable, open a public issue titled **Security contact
request** with no technical detail so a private channel can be arranged. Use synthetic values and never
post real wallet addresses, keys, seed phrases or credentials in a report.

> [card] [Evidence Model (WWEP)](https://docs.walletwall.org/mcp/evidence-model.md)
>
> See what WalletWall's evidence states do — and do not — claim.

> [card] [Troubleshooting](https://docs.walletwall.org/mcp/troubleshooting.md)
>
> Diagnose setup, input and resolver problems by symptom.

> [card] [Data Quality and Observability](https://docs.walletwall.org/architecture/data-quality.md)
>
> The hosted application's own provider boundary: what a wallet lookup sends to which provider, and how responses are labelled.

> [card] [Public and private repositories](https://docs.walletwall.org/concepts/public-and-private.md)
>
> Which WalletWall code is open to inspection, including the MCP engine, and what you cannot infer from it.


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