# TokenZero (remote · tokenzero.sh)

LLM cost optimization gateway, repo token-waste auditor & model arbitrage MCP server.

- Trust score: 77/100 (medium)
- Change this week: +5
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-11-04

## Components

- remote · `tokenzero.sh`: 77/100 (this document), [markdown](https://verifymcp.io/servers/tokenzero-ai-tokenzero/tokenzero.md), [page](https://verifymcp.io/servers/tokenzero-ai-tokenzero/tokenzero)

## Channel facts

- Endpoint: `https://tokenzero.sh/mcp`
- Transports: `streamable-http`
- Auth: `none`
- Version: `0.2.2`

## Trust breakdown

How this component scores in each security and reliability category. Every signal is checked automatically against the live server, and we only credit what we can confirm. Scores are 0–100 per category. Scoring method: https://verifymcp.io/docs/scoring (what has changed: https://verifymcp.io/docs/scoring/changelog)

Scored 2026-11-04.

- **Endpoint Security**: 80/100
  - The endpoint's TLS certificate is valid, in date, and uses a strong key.
  - No authorisation is required to call this server. Every tool declares its destructiveHint and none is destructive, so open access doesn't expose one.
  - HTTPS is enforced; there's no plaintext access path.
  - The HSTS (Strict-Transport-Security) header is present.
  - DNSSEC check failed: this domain isn't protected by DNSSEC.
- **Transport & Reachability**: 100/100
  - Verified streamable-http transport via a live MCP handshake.
- **Schema Quality & AI Usability**: 76/100
  - 100% of prompts and resources have a non-trivial description (not blank, and not just the item's name).
  - AI-judged instruction clarity (good).
  - Context-footprint check failed: tool/resource definitions use about 974 tokens (~162/item across 6 items; 6 tools + 0 resources), over budget; trim descriptions and params.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 30/100
  - Stability observed for 9 of 30 days with no destabilising changes; credit accrues until the full window elapses.
- **Tool Coverage**: 91/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 72% of tool parameters carry a description.
- **Tool Safety**: 100/100
  - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.
  - We read all 6 captured tool definition(s), and no name or description among them implies an irreversible operation.
  - An AI judge read all 7 captured unit(s) of tool text and found none that tries to manipulate the model reading it.
- **Capabilities**: 100/100
  - Implements a current MCP spec version (2026-07-28).

## Install

### How do I install the TokenZero MCP server?

TokenZero is a hosted endpoint at https://tokenzero.sh/mcp, so there is nothing to install locally. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.

### Claude

```bash
claude mcp add --transport http tokenzero-ai-tokenzero 'https://tokenzero.sh/mcp'
```

### Cursor

```json
{
  "mcpServers": {
    "tokenzero-ai-tokenzero": {
      "url": "https://tokenzero.sh/mcp"
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "tokenzero-ai-tokenzero": {
      "type": "http",
      "url": "https://tokenzero.sh/mcp"
    }
  }
}
```

### Codex

```toml
[mcp_servers.tokenzero-ai-tokenzero]
url = "https://tokenzero.sh/mcp"
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "tokenzero-ai-tokenzero": {
      "type": "remote",
      "url": "https://tokenzero.sh/mcp",
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add tokenzero-ai-tokenzero --url 'https://tokenzero.sh/mcp' --transport streamable-http
```

### Hermes

```yaml
mcp_servers:
  tokenzero-ai-tokenzero:
    url: "https://tokenzero.sh/mcp"
```

### Netclaw

```json
{
  "McpServers": {
    "tokenzero-ai-tokenzero": {
      "Transport": "http",
      "Url": "https://tokenzero.sh/mcp"
    }
  }
}
```

### Vellum

```bash
assistant mcp add tokenzero-ai-tokenzero -t streamable-http -u 'https://tokenzero.sh/mcp'
```

### Other

```json
{
  "mcpServers": {
    "tokenzero-ai-tokenzero": {
      "type": "http",
      "url": "https://tokenzero.sh/mcp"
    }
  }
}
```

The mcpServers block is a cross-client convention. Remote transports vary, so check your client's docs.

## Changelog

Every change recorded for this component, newest first. Days that predate change tracking, or that we cannot explain, say so: "we were watching and nothing happened" and "we were not watching" are different claims.

### 2026-11-04 (score 77, 0)

- [functional] Server version: 0.2.15 → 0.2.16

### 2026-11-03 (score 77, +1)

- [functional] Server version: 0.2.12 → 0.2.15

### 2026-11-02 (score 76, 0)

- [functional] Server version: 0.2.11 → 0.2.12

### 2026-11-01 (score 76, +3)

- [security] The server rewrote its instructions, which are the text every model session reads
- [security] Tool “calculate_model_arbitrage” rewrote its description, which is the text the model reads
- [functional regression] Schema quality: 781 → 974
- [functional improvement] Tool coverage: 67% → 72%
- [functional] The server now declares the “prompts” capability
- [functional] First check of Schema quality: 100
- [functional] New prompt “cut-my-llm-bill”
- [functional] New prompt “price-my-workload”
- [functional] New prompt “try-tokenzero”
- [functional] Server version: 0.2.9 → 0.2.11
- [functional] New tool “try_tokenzero”

### 2026-10-31 (score 73, +1)

No change was recorded against any check on this day. Stability & Change Management went from 13 to 17. That category is still filling its 30-day observation window: 4 days of observed history at the previous scan, 5 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-10-29 (score 72, 0)

- [security] Tool “audit_repo_token_waste” rewrote its description, which is the text the model reads
- [security] Tool “calculate_model_arbitrage” rewrote its description, which is the text the model reads
- [security] Tool “optimize_prompt_payload” rewrote its description, which is the text the model reads
- [functional regression] Schema quality: 129 → 156
- [functional improvement] Tool coverage: 47% → 67%
- [functional] Server version: 0.2.8 → 0.2.9
- [cosmetic] “calculate_model_arbitrage” reworded the description of “input_tokens”
- [cosmetic] “calculate_model_arbitrage” reworded the description of “model”
- [cosmetic] “calculate_model_arbitrage” reworded the description of “output_tokens”
- [cosmetic] “calculate_model_arbitrage” made “input_tokens” optional
- [cosmetic] “calculate_model_arbitrage” made “output_tokens” optional

### 2026-10-28 (score 72, +1)

- [security] Tool “calculate_model_arbitrage” rewrote its description, which is the text the model reads
- [functional] Server version: 0.2.7 → 0.2.8

### 2026-10-27 (score 71, +7)

- [security improvement] Authorization: unverified → partial
- [security] The server rewrote its instructions, which are the text every model session reads
- [functional regression] Schema quality: pass → fail
- [functional improvement] Stability: unverified → 0.03
- [functional] Server version: 0.2.2 → 0.2.7
- [cosmetic] Tool “audit_repo_token_waste” changed its title: Audit a repo for LLM token waste
- [cosmetic] Tool “calculate_model_arbitrage” changed its title: Price a workload across models
- [cosmetic] Tool “optimize_prompt_payload” changed its title: Make a request body cache-friendly
- [cosmetic] Tool “provision_workspace” changed its title: Create a free TokenZero workspace
- [cosmetic] Tool “workspace_metrics” changed its title: Read a workspace savings scorecard

## MCP tools (6)

### `try_tokenzero` (~169 tokens)

Try TokenZero (no key needed)

Call this first: no API key, no signup, no setup. Shows what TokenZero would do to one of your LLM calls: it runs the four savings engines (prompt-cache formatting, edge cache, cascade advice, signed receipt) on a sample or your own prompt and model, and returns the list-price saving, the signed receipt and a copy-paste keyless curl that makes a real call on a free model. Pass {} for the default, or {"model": "gpt-6.1-sol", "system": "...", "prompt": "..."}.

Input parameters:

- `model` (string): a priced model id; default gpt-6.1-sol
- `prompt` (string): your user message (optional)
- `system` (string): your system prompt (optional)

### `audit_repo_token_waste` (~149 tokens)

Audit a repo for LLM token waste

No API key needed. Static-analyze a public GitHub repo (e.g. {"repo": "owner/name"}) or a set of source files for LLM token waste: uncached static prompts, frontier models on extraction tasks, hot loops, non-deterministic repeat calls, and prompt-caching mis-integration (caching gated on model names, volatile values in the system prompt, random ids in history, wrong-provider cache markers, misplaced breakpoints, cost math that ignores cached tokens). Returns findings with fixes and an estimated recoverable percentage.

Input parameters:

- `files` (array): Alternative to repo: files to analyze
- `repo` (string): GitHub owner/name (public)

### `optimize_prompt_payload` (~99 tokens)

Make a request body cache-friendly

No API key needed. Rewrite an OpenAI Chat/Responses or Anthropic Messages request body so its static prefix is provider-cacheable (Anthropic cache_control breakpoints, OpenAI prompt_cache_key). Returns the optimized body, what changed, the token profile and the projected per-request saving. Never changes the tokens the model sees.

Input parameters:

- `body` (object, required): The request body
- `endpoint` (string)
- `strategy` (string)

### `calculate_model_arbitrage` (~187 tokens)

Price a workload across models

No API key needed; only `model` is required (e.g. {"model": "gpt-6.1-sol"}, defaults 2,000 tokens in and 400 out). Price a workload at list prices across models: cost per request and per month for the requested model, its cheaper sibling, and the effect of prompt caching. The result names the pricing table version it used.

Input parameters:

- `compare_to` (array)
- `input_tokens` (number): per request; default 2000
- `model` (string, required): model id, e.g. gpt-6.1-sol or claude-sonnet-5-5
- `output_tokens` (number): per request; default 400
- `requests_per_month` (number)
- `static_tokens` (number): portion of input_tokens that is a fixed prefix (system + tools)

### `provision_workspace` (~80 tokens)

Create a free TokenZero workspace

Create a TokenZero workspace and key with no form. Returns the key once, the base URLs that route through it (OPENAI_BASE_URL / ANTHROPIC_BASE_URL), the one-click dashboard link and the badge URL. Free up to the monthly free tier; nothing is billed.

Input parameters:

- `name` (string): workspace name (default mcp-agent)

### `workspace_metrics` (~102 tokens)

Read a workspace savings scorecard

Live value scorecard for a TokenZero workspace: hard savings and net ROI per window, the four engines' contribution, speed, parity, availability and privacy posture, per-model breakdown and one-click recommendations. Needs the workspace key (tz_live_…); pass demo: true for a labelled sample.

Input parameters:

- `demo` (boolean): return the sample scorecard instead
- `key` (string): workspace key (tz_live_…)
- `window` (string)

## Diagnostics

Captured diagnostic sections: TLS, DNSSEC, Authorisation, Transports. The full working is on the page: https://verifymcp.io/servers/tokenzero-ai-tokenzero/tokenzero#diagnostics

## Score history

- 2026-11-04: 77
- 2026-11-03: 77
- 2026-11-02: 76
- 2026-11-01: 76
- 2026-10-31: 73
- 2026-10-30: 72
- 2026-10-29: 72
- 2026-10-28: 72
- 2026-10-27: 71
- 2026-10-26: 64

## Common questions

### What is the TokenZero MCP server?

TokenZero is an MCP server listed in the public MCP registry as io.github.tokenzero-ai/tokenzero. LLM cost optimization gateway, repo token-waste auditor & model arbitrage MCP server. This page covers its hosted endpoint (https://tokenzero.sh/mcp).

### Is the TokenZero MCP server safe to use?

TokenZero scores 77 out of 100 on VerifyMCP. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.

### What tools does the TokenZero MCP server expose?

TokenZero exposes 6 tools: try_tokenzero, audit_repo_token_waste, optimize_prompt_payload, calculate_model_arbitrage, provision_workspace, workspace_metrics. Their descriptions and schemas cost roughly 786 tokens of context every time the server is loaded.

### Does the TokenZero MCP server require authentication?

No. We connected to TokenZero without credentials and it answered, so anything it exposes is reachable by anyone who knows the address.

### Is the TokenZero MCP server still maintained?

TokenZero is still listed as active in the MCP registry. We last reached this channel on 4 November 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.

## Links

- Remote endpoint: https://tokenzero.sh/mcp
- Repository: https://github.com/tokenzero-ai/tokenzero
- Website: https://tokenzero.sh/
- Changelog RSS feed: https://verifymcp.io/servers/tokenzero-ai-tokenzero/tokenzero.xml
- Changelog JSON feed: https://verifymcp.io/servers/tokenzero-ai-tokenzero/tokenzero.json
- HTML version of this page: https://verifymcp.io/servers/tokenzero-ai-tokenzero/tokenzero