# io.github.robertoyanguas-ops/agentdiet (remote · agentdiet-72343955726.us-central1.run.app)

Strips 85-97% of HTML bloat into clean Markdown, <200-token skeletons, or BM25 excerpts via MCP.

- Trust score: 68/100 (medium)
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-10-18

## Components

- remote · `agentdiet-72343955726.us-central1.run.app`: 68/100 (this document), [markdown](https://verifymcp.io/servers/robertoyanguas-ops-agentdiet/agentdiet-72343955726.md), [page](https://verifymcp.io/servers/robertoyanguas-ops-agentdiet/agentdiet-72343955726)

## Channel facts

- Endpoint: `https://agentdiet-72343955726.us-central1.run.app/mcp`
- Transports: `streamable-http`
- Auth: `none`
- Version: `1.1.1`

## 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 2026-10-18.

- **Endpoint Security**: 74/100
  - 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.
  - HSTS check failed: the Strict-Transport-Security header is absent.
  - 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**: 72/100
  - 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 1547 tokens (~257/item across 6 items; 4 tools + 2 resources), over budget; trim descriptions and params.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 0/100
  - Stability check failed: schema churn in the 5 days we've observed: 15 tool removals, 0 breaking changes, 0 auth/transport breaks, 0 additions.
- **Tool Coverage**: 100/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 100% of tool parameters carry a description.
  - Structured output schemas are declared (100% of tools); any adoption earns full credit.
- **Tool Safety**: 100/100
  - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.
  - We read all 4 captured tool definition(s), and no name or description among them implies an irreversible operation.
  - An AI judge read all 6 captured unit(s) of tool text and found none that tries to manipulate the model reading it.
- **Capabilities**: 40/100
  - Spec-recency check failed: implements MCP spec 2025-03-26; the latest is 2026-07-28.

## Install

### How do I install the io.github.robertoyanguas-ops/agentdiet MCP server?

io.github.robertoyanguas-ops/agentdiet is a hosted endpoint at https://agentdiet-72343955726.us-central1.run.app/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 robertoyanguas-ops-agentdiet 'https://agentdiet-72343955726.us-central1.run.app/mcp'
```

### Cursor

```json
{
  "mcpServers": {
    "robertoyanguas-ops-agentdiet": {
      "url": "https://agentdiet-72343955726.us-central1.run.app/mcp"
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "robertoyanguas-ops-agentdiet": {
      "type": "http",
      "url": "https://agentdiet-72343955726.us-central1.run.app/mcp"
    }
  }
}
```

### Codex

```toml
[mcp_servers.robertoyanguas-ops-agentdiet]
url = "https://agentdiet-72343955726.us-central1.run.app/mcp"
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "robertoyanguas-ops-agentdiet": {
      "type": "remote",
      "url": "https://agentdiet-72343955726.us-central1.run.app/mcp",
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add robertoyanguas-ops-agentdiet --url 'https://agentdiet-72343955726.us-central1.run.app/mcp' --transport streamable-http
```

### Hermes

```yaml
mcp_servers:
  robertoyanguas-ops-agentdiet:
    url: "https://agentdiet-72343955726.us-central1.run.app/mcp"
```

### Netclaw

```json
{
  "McpServers": {
    "robertoyanguas-ops-agentdiet": {
      "Transport": "http",
      "Url": "https://agentdiet-72343955726.us-central1.run.app/mcp"
    }
  }
}
```

### Vellum

```bash
assistant mcp add robertoyanguas-ops-agentdiet -t streamable-http -u 'https://agentdiet-72343955726.us-central1.run.app/mcp'
```

### Other

```json
{
  "mcpServers": {
    "robertoyanguas-ops-agentdiet": {
      "type": "http",
      "url": "https://agentdiet-72343955726.us-central1.run.app/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.

### 2026-10-18 (score 68, 0)

- [functional improvement] Tool “distill_webpage” now declares an output schema
- [functional improvement] Tool “extract_structured_data” now declares an output schema
- [functional improvement] Tool “get_agent_wallet_status” now declares an output schema
- [functional improvement] Tool “recon_webpage_skeleton” now declares an output schema
- [functional] First check of Tool coverage: 100
- [cosmetic] “get_agent_wallet_status” dropped the optional parameter “tx_hash”

### 2026-10-17 (score 68, −1)

- [security] The server rewrote its instructions, which are the text every model session reads
- [security] Tool “distill_webpage” rewrote its description, which is the text the model reads
- [security] Tool “extract_structured_data” rewrote its description, which is the text the model reads
- [security] Tool “get_agent_wallet_status” rewrote its description, which is the text the model reads
- [security] Tool “recon_webpage_skeleton” rewrote its description, which is the text the model reads
- [functional regression] Schema quality: pass → fail
- [functional improvement] Tool coverage: 80% → 100%
- [cosmetic] “distill_webpage” added an optional parameter “agent_id”
- [cosmetic] “distill_webpage” added an optional parameter “tx_hash”
- [cosmetic] “extract_structured_data” added an optional parameter “agent_id”
- [cosmetic] “extract_structured_data” added an optional parameter “tx_hash”
- [cosmetic] “get_agent_wallet_status” added an optional parameter “tx_hash”
- [cosmetic] “recon_webpage_skeleton” added an optional parameter “agent_id”
- [cosmetic] “recon_webpage_skeleton” added an optional parameter “tx_hash”
- [cosmetic] “distill_webpage” reworded the description of “html”
- [cosmetic] “distill_webpage” reworded the description of “max_tokens”
- [cosmetic] “distill_webpage” reworded the description of “mode”
- [cosmetic] “distill_webpage” reworded the description of “query”
- [cosmetic] “distill_webpage” reworded the description of “url”
- [cosmetic] “extract_structured_data” reworded the description of “html”
- [cosmetic] “extract_structured_data” reworded the description of “url”
- [cosmetic] “get_agent_wallet_status” reworded the description of “agent_id”
- [cosmetic] “recon_webpage_skeleton” reworded the description of “html”
- [cosmetic] “recon_webpage_skeleton” reworded the description of “url”

### 2026-10-16 (score 69, +5)

- [security regression] Stability: 0.07 → fail
- [security regression] Tool “audit_github_repository” was removed
- [security regression] Tool “audit_x402_and_a2a_endpoint” was removed
- [security regression] Tool “discover_and_verify_x402_mcp_tools” was removed
- [security regression] Tool “enrich_company_domain” was removed
- [security regression] Tool “extract_ecommerce_catalog” was removed
- [security regression] Tool “fetch_arxiv_research_digest” was removed
- [security regression] Tool “generate_site_llms_txt” was removed
- [security regression] Tool “get_dex_token_intelligence” was removed
- [security regression] Tool “get_public_company_sec_intel” was removed
- [security regression] Tool “recon_domain_osint” was removed
- [security regression] Tool “repair_malformed_json” was removed
- [security regression] Tool “scan_onchain_address_risk” was removed
- [security regression] Tool “scan_prompt_injection” was removed
- [security regression] Tool “search_clean_agent_web” was removed
- [security regression] Tool “verify_email_deliverability” was removed
- [security improvement] Authorization: unverified → partial
- [security] Tool “distill_webpage” rewrote its description, which is the text the model reads
- [security] Tool “extract_structured_data” rewrote its description, which is the text the model reads
- [security] Tool “get_agent_wallet_status” rewrote its description, which is the text the model reads
- [security] Tool “recon_webpage_skeleton” rewrote its description, which is the text the model reads
- [functional regression] Tool coverage: 92% → 80%
- [functional regression] Schema quality: 76 → 104
- [functional] Server version: 1.0.0 → 6.3.0

### 2026-10-15 (score 64, 0)

- [security] Tool “audit_github_repository” rewrote its description, which is the text the model reads
- [security] Tool “audit_x402_and_a2a_endpoint” rewrote its description, which is the text the model reads
- [security] Tool “discover_and_verify_x402_mcp_tools” rewrote its description, which is the text the model reads
- [security] Tool “enrich_company_domain” rewrote its description, which is the text the model reads
- [security] Tool “extract_ecommerce_catalog” rewrote its description, which is the text the model reads
- [security] Tool “fetch_arxiv_research_digest” rewrote its description, which is the text the model reads
- [security] Tool “generate_site_llms_txt” rewrote its description, which is the text the model reads
- [security] Tool “get_dex_token_intelligence” rewrote its description, which is the text the model reads
- [security] Tool “get_public_company_sec_intel” rewrote its description, which is the text the model reads
- [security] Tool “repair_malformed_json” rewrote its description, which is the text the model reads
- [security] Tool “scan_onchain_address_risk” rewrote its description, which is the text the model reads
- [security] Tool “scan_prompt_injection” rewrote its description, which is the text the model reads
- [security] Tool “search_clean_agent_web” rewrote its description, which is the text the model reads
- [security] Tool “verify_email_deliverability” rewrote its description, which is the text the model reads
- [cosmetic] “audit_github_repository” reworded the description of “repo”
- [cosmetic] “audit_x402_and_a2a_endpoint” reworded the description of “url”
- [cosmetic] “discover_and_verify_x402_mcp_tools” reworded the description of “query”
- [cosmetic] “generate_site_llms_txt” reworded the description of “url”
- [cosmetic] “get_dex_token_intelligence” reworded the description of “token”
- [cosmetic] “get_public_company_sec_intel” reworded the description of “ticker”
- [cosmetic] “recon_domain_osint” reworded the description of “domain”
- [cosmetic] “scan_onchain_address_risk” reworded the description of “address”
- [cosmetic] “scan_prompt_injection” reworded the description of “text”

### 2026-10-14 (score 64, +1)

- [functional improvement] Stability: unverified → 0.03

### 2026-10-13 (score 63)

First indexed and scored.

## MCP tools (4)

### `distill_webpage` (~480 tokens)

Fetch a public web page (or take raw html) and return its main text as clean Markdown, removing 85-97% of HTML/JS/CSS boilerplate. mode=markdown returns the whole cleaned page; mode=budget returns only the sections most relevant to query, capped at max_tokens. Use it to read page content. To preview headings and size first, use recon_webpage_skeleton; for JSON-LD, tables or code blocks, use extract_structured_data. Not for PDFs, images or login-only pages. Read-only; results are cached for 1 hour. $0.001 per call ($0.0005 when cached). Returns JSON: data.content (Markdown), data.metadata, and roi_proof (tokens before/after). Pricing: the first 5 calls per agent_id are free and failed calls are never charged. After that the tool returns a payment-required message (not an error) with a Stripe top-up link (from $2) and Base/Solana USDC addresses; pay, then repeat the same call with the same agent_id, or add tx_hash after a USDC transfer.

Input parameters:

- `agent_id` (string): Your agent ID, or the agd_pass_ key you got from a payment. It tracks your free calls and credit, so treat it like a password. Omit it to use an ID derived from your IP and client.
- `html` (string): Raw HTML (up to ~1.9MB) to process instead of fetching url. When set, url only labels the cache entry.
- `max_tokens` (integer): Output token cap for mode=budget (50-16000, default 400).
- `mode` (string): markdown (default) = whole page; budget = only sections relevant to query.
- `query` (string): Keywords used to rank sections (BM25). Used only when mode=budget.
- `tx_hash` (string): Use only after paying on-chain: the Base (0x...) or Solana USDC transfer hash. It is verified on-chain before the call runs, and any overpayment is returned as an agd_pass_ credit key.
- `url` (string, required): Public http(s) URL to fetch, or builtin://stripe-docs for a demo page. Private and metadata IPs are blocked.

Output parameters:

- `data` (object)
- `roi_proof` (object)
- `status` (string)

### `recon_webpage_skeleton` (~363 tokens)

Return a compact outline (under ~200 tokens) of a web page: its H1-H4 headings with a token count per section. Call it first on long or unfamiliar pages to decide whether reading is worth it, then call distill_webpage (with mode=budget and a query to read just the relevant part). It returns no body text. Read-only; results are cached for 1 hour. $0.0002 per call. Returns JSON: data.outline, data.summary_stats, data.metadata. Pricing: the first 5 calls per agent_id are free and failed calls are never charged. After that the tool returns a payment-required message (not an error) with a Stripe top-up link (from $2) and Base/Solana USDC addresses; pay, then repeat the same call with the same agent_id, or add tx_hash after a USDC transfer.

Input parameters:

- `agent_id` (string): Your agent ID, or the agd_pass_ key you got from a payment. It tracks your free calls and credit, so treat it like a password. Omit it to use an ID derived from your IP and client.
- `html` (string): Raw HTML (up to ~1.9MB) to process instead of fetching url. When set, url only labels the cache entry.
- `tx_hash` (string): Use only after paying on-chain: the Base (0x...) or Solana USDC transfer hash. It is verified on-chain before the call runs, and any overpayment is returned as an agd_pass_ credit key.
- `url` (string, required): Public http(s) URL to fetch, or builtin://stripe-docs for a demo page. Private and metadata IPs are blocked.

Output parameters:

- `data` (object)
- `roi_proof` (object)
- `status` (string)

### `extract_structured_data` (~368 tokens)

Extract machine-readable data from a web page: Schema.org JSON-LD, HTML tables as JSON rows, code blocks, section list and links. Use it for prices, specs, product or API tables and code samples. For prose, use distill_webpage; for a heading outline, use recon_webpage_skeleton. Read-only; results are cached for 1 hour. $0.001 per call ($0.0005 when cached). Returns JSON: data.json_ld, data.tables, data.code_blocks, data.sections, data.links. Pricing: the first 5 calls per agent_id are free and failed calls are never charged. After that the tool returns a payment-required message (not an error) with a Stripe top-up link (from $2) and Base/Solana USDC addresses; pay, then repeat the same call with the same agent_id, or add tx_hash after a USDC transfer.

Input parameters:

- `agent_id` (string): Your agent ID, or the agd_pass_ key you got from a payment. It tracks your free calls and credit, so treat it like a password. Omit it to use an ID derived from your IP and client.
- `html` (string): Raw HTML (up to ~1.9MB) to process instead of fetching url. When set, url only labels the cache entry.
- `tx_hash` (string): Use only after paying on-chain: the Base (0x...) or Solana USDC transfer hash. It is verified on-chain before the call runs, and any overpayment is returned as an agd_pass_ credit key.
- `url` (string, required): Public http(s) URL to fetch, or builtin://stripe-docs for a demo page. Private and metadata IPs are blocked.

Output parameters:

- `data` (object)
- `roi_proof` (object)
- `status` (string)

### `get_agent_wallet_status` (~141 tokens)

Show an agent's account: free calls used (of 5), prepaid USD balance, total spent, total calls and tokens saved. Free: it never charges or uses a free call. Call it before a batch of paid calls, or after a top-up to confirm the credit arrived. It does not process content or take payments; use the other tools on this server for that. Returns JSON with free_calls_used, balance_usd, total_spent_usd, total_tokens_saved.

Input parameters:

- `agent_id` (string): Agent ID or agd_pass_ key to look up. Omit it to see the ID derived from your IP and client.

Output parameters:

- `data` (object)
- `roi_proof` (object)
- `status` (string)

## Diagnostics

Captured diagnostic sections: TLS, DNSSEC, Authorisation, Transports. The full working is on the page: https://verifymcp.io/servers/robertoyanguas-ops-agentdiet/agentdiet-72343955726#diagnostics

## Score history

- 2026-10-18: 68
- 2026-10-17: 68
- 2026-10-16: 69
- 2026-10-15: 64
- 2026-10-14: 64
- 2026-10-13: 63

## Common questions

### What is the io.github.robertoyanguas-ops/agentdiet MCP server?

io.github.robertoyanguas-ops/agentdiet is an MCP server listed in the public MCP registry as io.github.robertoyanguas-ops/agentdiet. Strips 85-97% of HTML bloat into clean Markdown, <200-token skeletons, or BM25 excerpts via MCP. This page covers its hosted endpoint (https://agentdiet-72343955726.us-central1.run.app/mcp).

### Is the io.github.robertoyanguas-ops/agentdiet MCP server safe to use?

io.github.robertoyanguas-ops/agentdiet scores 68 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 io.github.robertoyanguas-ops/agentdiet MCP server expose?

io.github.robertoyanguas-ops/agentdiet exposes 4 tools: distill_webpage, recon_webpage_skeleton, extract_structured_data, get_agent_wallet_status. Their descriptions and schemas cost roughly 1,352 tokens of context every time the server is loaded.

### Does the io.github.robertoyanguas-ops/agentdiet MCP server require authentication?

No. We connected to io.github.robertoyanguas-ops/agentdiet without credentials and it answered, so anything it exposes is reachable by anyone who knows the address.

### Is the io.github.robertoyanguas-ops/agentdiet MCP server still maintained?

io.github.robertoyanguas-ops/agentdiet is still listed as active in the MCP registry. We last reached this channel on 18 October 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://agentdiet-72343955726.us-central1.run.app/mcp
- Changelog RSS feed: https://verifymcp.io/servers/robertoyanguas-ops-agentdiet/agentdiet-72343955726.xml
- Changelog JSON feed: https://verifymcp.io/servers/robertoyanguas-ops-agentdiet/agentdiet-72343955726.json
- HTML version of this page: https://verifymcp.io/servers/robertoyanguas-ops-agentdiet/agentdiet-72343955726