# ComplyOnSite (remote · mcp.complyonsite.com)

Waste codes, transfer notes filled in from chat, HAVS, noise and first-aid calculators.

- Trust score: 82/100 (high trust)
- Change this week: +1
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-11-05

## Components

- remote · `mcp.complyonsite.com`: 82/100 (this document), [markdown](https://verifymcp.io/servers/com-complyonsite-tools/mcp.md), [page](https://verifymcp.io/servers/com-complyonsite-tools/mcp)

## Channel facts

- Endpoint: `https://mcp.complyonsite.com/mcp`
- Transports: `streamable-http`
- Auth: `none`
- Version: `0.4.0`

## 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 2026-11-05.

- **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**: 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 2339 tokens (~179/item across 13 items; 11 tools + 2 resources), over budget; trim descriptions and params.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 93/100
  - Stability observed for 28 of 30 days with no destabilising changes; credit accrues until the full window elapses.
- **Tool Coverage**: 96/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 86% of tool parameters carry a description.
  - Structured output schemas are declared (100% of tools); any adoption earns full credit.
- **Tool Safety**: 75/100
  - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.
  - 0 of 1 tool(s) whose name or description implies an irreversible operation declare an MCP destructiveHint annotation; "fill_transfer_note" implies "transfer" and declares readOnlyHint instead, contradicting what its own name says it does.
  - An AI judge read all 13 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).
  - Supports UI / widget rendering.

## Install

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

ComplyOnSite is a hosted endpoint at https://mcp.complyonsite.com/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 com-complyonsite-tools 'https://mcp.complyonsite.com/mcp'
```

### Cursor

```json
{
  "mcpServers": {
    "com-complyonsite-tools": {
      "url": "https://mcp.complyonsite.com/mcp"
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "com-complyonsite-tools": {
      "type": "http",
      "url": "https://mcp.complyonsite.com/mcp"
    }
  }
}
```

### Codex

```toml
[mcp_servers.com-complyonsite-tools]
url = "https://mcp.complyonsite.com/mcp"
```

### opencode

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

### OpenClaw

```bash
openclaw mcp add com-complyonsite-tools --url 'https://mcp.complyonsite.com/mcp' --transport streamable-http
```

### Hermes

```yaml
mcp_servers:
  com-complyonsite-tools:
    url: "https://mcp.complyonsite.com/mcp"
```

### Netclaw

```json
{
  "McpServers": {
    "com-complyonsite-tools": {
      "Transport": "http",
      "Url": "https://mcp.complyonsite.com/mcp"
    }
  }
}
```

### Vellum

```bash
assistant mcp add com-complyonsite-tools -t streamable-http -u 'https://mcp.complyonsite.com/mcp'
```

### Other

```json
{
  "mcpServers": {
    "com-complyonsite-tools": {
      "type": "http",
      "url": "https://mcp.complyonsite.com/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.

### 2026-11-03 (score 82, +1)

No change was recorded against any check on this day. Stability & Change Management went from 83 to 87. That category is still filling its 30-day observation window: 25 days of observed history at the previous scan, 26 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-11-01 (score 81, −1)

- [security] Resource “result-card” now points somewhere else: ui://complyonsite/result-v2.html → ui://complyonsite/result-v3.html
- [security] The server rewrote its instructions, which are the text every model session reads
- [functional regression] Tool coverage: 100% → 86%
- [functional regression] Schema quality: 1689 → 2339
- [functional] Destructive annotations: pass → 0
- [functional] New resource “held-result-v2.html”
- [functional] Server version: 0.3.0 → 0.4.0
- [functional] New tool “fill_transfer_note”
- [functional] New tool “site_tools”
- [functional] New tool “waste_code_search”

### 2026-10-30 (score 82, +1)

No change was recorded against any check on this day. Stability & Change Management went from 70 to 73. That category is still filling its 30-day observation window: 21 days of observed history at the previous scan, 22 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-10-28 (score 81, +1)

No change was recorded against any check on this day. Stability & Change Management went from 63 to 67. That category is still filling its 30-day observation window: 19 days of observed history at the previous scan, 20 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-10-26 (score 80, +1)

No change was recorded against any check on this day. Stability & Change Management went from 57 to 60. That category is still filling its 30-day observation window: 17 days of observed history at the previous scan, 18 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-10-24 (score 79, +1)

No change was recorded against any check on this day. Stability & Change Management went from 50 to 53. That category is still filling its 30-day observation window: 15 days of observed history at the previous scan, 16 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-10-22 (score 78, +1)

No change was recorded against any check on this day. Stability & Change Management went from 43 to 47. That category is still filling its 30-day observation window: 13 days of observed history at the previous scan, 14 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-10-20 (score 77, +1)

No change was recorded against any check on this day. Stability & Change Management went from 37 to 40. That category is still filling its 30-day observation window: 11 days of observed history at the previous scan, 12 at this one. The score rises as the window fills, whether or not the server changes.

## MCP tools (11)

### `ewc_search` (~200 tokens)

Find UK waste codes

Find UK List of Waste (EWC) codes for a waste, skip or load from one to three material words such as “plasterboard”, “mixed rubble” or “contaminated soil”, from one or more codes, or from a code prefix. Search each material separately, not a sentence; match says how many words the result covers, and a weak match returns no waste. Returns official entries with hazardous and mirror status, partner codes, source lineage and any assessment needed. Use the returned entries as candidates: origin, process and composition still determine the classification. To start a note with chosen codes, pass them to start_document.

Input parameters:

- `limit` (integer): Maximum matched waste groups and official entries to return; defaults to 5.
- `query` (string, required): One to three material words, one or more EWC/LoW codes, or a code prefix; common spaced, hyphenated and unspaced formats are accepted.

Output parameters:

- `limit` (string): What it does not establish.
- `source` (object)
- `summary` (string): The answer, figure or code first.

### `first_aid_requirements` (~207 tokens)

Estimate first-aid cover

Estimate first-aid cover for each shift from HSE guidance: how many first-aiders or appointed persons, which course, whether the first-aiders already there are enough, and whether a first-aid room is worth having. Construction site by default; also factory, office, shop, nursery (EYFS) and school. Returns roles, training, shortages, practical steps, factors, source lineage and a calculator link with the same inputs. This starts a workplace needs assessment; it is not a statutory staffing quota.

Input parameters:

- `factors` (array): Relevant factor ids returned by this tool for the selected workplace.
- `firstAidersAvailable` (array): First-aiders already available on each shift, in the same order.
- `receptionClass` (boolean): For a school: whether it has a reception or nursery class.
- `shifts` (array, required): Highest number of people present at once on each shift.
- `work` (string): Workplace setting; defaults to construction site.

Output parameters:

- `limit` (string): What it does not establish.
- `source` (object)
- `summary` (string): The answer, figure or code first.

### `havs_daily_exposure` (~164 tokens)

Calculate daily HAVS exposure

Calculate one person’s daily hand-arm vibration (HAVS) exposure from the tools they use and for how long: A(8), HSE points, low/high range, and whether it passes the action value (2.5 m/s² A(8), 100 points) or the limit value (5 m/s², 400 points). For each tool give its own vibration figure in m/s², or an HSE category and variant to use HSE’s typical figure and range. Time means trigger time: the tool running in the person’s hands. Returns a link that opens the same calculation in the HAVS calculator.

Input parameters:

- `tools` (array, required): Tools or tasks used by one person during one day; 1 to 20 entries.

Output parameters:

- `limit` (string): What it does not establish.
- `source` (object)
- `summary` (string): The answer, figure or code first.

### `noise_daily_exposure` (~233 tokens)

Calculate daily noise exposure

Calculate one person’s daily noise exposure (LEP,d) from measured task levels (LAeq in dB(A)) and how long each lasts, or from dosemeter readings, and which action value it reaches (80 and 85 dB(A)). Optionally gives the weekly LEP,w, compares peak dB(C), and estimates noise at the ear against the 87 dB(A) limit from hearing protection SNR or H/M/L with each task’s C-weighted level. Do not use sound power LwA or per-shot values; omitted peak or protection measurements remain explicitly unassessed. Returns a link that opens the same calculation in the noise calculator.

Input parameters:

- `peak` (number): Highest measured C-weighted peak sound pressure level in dB(C).
- `protector`: Hearing-protector attenuation from its SNR or H/M/L ratings.
- `tasks` (array, required): Measured tasks for one person in one day; 1 to 10 entries.
- `week`: Repeat the calculated day this many days, or supply the daily LEP,d for each exposed day.

Output parameters:

- `limit` (string): What it does not establish.
- `source` (object)
- `summary` (string): The answer, figure or code first.

### `construction_injury_statistics` (~120 tokens)

Get construction injury statistics

Return a retained HSE series for construction in Great Britain: worker deaths and the rate per 100,000 workers with the long-run trend, kinds of fatal accident such as falls from height, employee deaths, self-reported and RIDDOR-reported injuries, or work-related ill health. Preserves each series’ period, units, provisional or revised status, confidence intervals, source file and publication date. This does not query a live HSE feed.

Input parameters:

- `series` (string): Dataset series to return; defaults to the latest headline figures.

Output parameters:

- `limit` (string): What it does not establish.
- `source` (object)
- `summary` (string): The answer, figure or code first.

### `power_tool_types` (~122 tokens)

Get power-tool exposure data

Return how much a type of power tool vibrates and how loud makers declare it, such as an angle grinder, SDS drill or breaker: the median and spread of declared figures across models, the matching HSE vibration category and variants, and HSE in-use noise comparisons where available. Omit type to list the types. It does not return maker-model rows or quotes; declared LpA is not measured task LAeq.

Input parameters:

- `type` (string): Tool-type slug returned by this tool, for example angle-grinder; omit to list all available types.

Output parameters:

- `limit` (string): What it does not establish.
- `source` (object)
- `summary` (string): The answer, figure or code first.

### `waste_paperwork` (~219 tokens)

Check waste paperwork requirements

Return which paperwork a waste movement needs: in the UK from the nation and whether the waste is hazardous, or any listed document by id, including Ireland, the Netherlands, the US, Canada and Australia. A document id takes precedence when both forms are supplied. Results cover the document, parties, the items the preparation tool asks for, retention, pre-notification where it applies, electronic and signature status, regulator route, the ComplyOnSite tool that prepares it, and source lineage. Preparation tools do not issue, submit or confirm acceptance of official documents.

Input parameters:

- `detail` (boolean): With document: also return the form’s full field, guide and FAQ specification. Large; only when filling the form in.
- `document` (string): A document id returned by this tool, for example us-asbestos-wsr or sct-special-waste.
- `hazardous` (boolean): Whether the waste is hazardous; use together with jurisdiction.
- `jurisdiction` (string): UK jurisdiction; use together with hazardous, or omit both to list documents.

Output parameters:

- `limit` (string): What it does not establish.
- `source` (object)
- `summary` (string): The answer, figure or code first.

### `start_document` (~99 tokens)

Get a compliance document link

Return an anonymous complyonsite.com link that opens a waste transfer note, hazardous waste consignment note or RAMS ready to fill in, plus the blank template and a worked example as ready-made PDFs. Waste-note links prefill validated EWC codes; RAMS links select a job and optional supported operation. The person completes, signs and downloads the document on the site. The tool does not create, save, sign, issue or submit a document.

Output parameters:

- `limit` (string): What it does not establish.
- `source` (object)
- `summary` (string): The answer, figure or code first.

### `fill_transfer_note` (~297 tokens)

Fill in a waste transfer note

Fill in a UK waste transfer note from what the person has said. Returns what the law still needs before signing, a card to correct every detail, and a link that opens the note unsigned on complyonsite.com, where both sides sign and save the PDF. Pass only details the person gave; never invent a name, address or registration number. Call again with the same details plus any new ones. Non-hazardous codes only; a hazardous load uses start_document. Saves, signs and sends nothing.

Input parameters:

- `codes` (array): Non-hazardous EWC codes from ewc_search, up to 5.
- `container` (string)
- `containerKind` (string)
- `date` (string): YYYY-MM-DD.
- `description` (string)
- `destination` (string): Where it goes after collection, for Northern Ireland.
- `from` (object): Who hands the waste over.
- `place` (object)
- `placeSame` (boolean): True when handed over at the from address.
- `quantity` (string): For example “1 × 8 yard skip” or “2 tonnes”.
- `region` (string): Omit if unknown: England is assumed and the result says so.
- `sic` (string): Five-digit SIC 2007 code of the person handing over.
- `time` (string): HH:MM.
- `to` (object): Who collects it.

Output parameters:

- `limit` (string): What it does not establish.
- `source` (object)
- `summary` (string): The answer, figure or code first.

### `waste_code_search` (~105 tokens)

Find waste codes outside the UK

Find waste codes in the US federal hazardous waste list (RCRA), California, Texas, Australia or Canada from one to three material words or a code. Returns the code, plain name, official wording where published, and what decides it. Use ewc_search for the UK. Candidates only, not a classification.

Input parameters:

- `limit` (integer)
- `query` (string, required): One to three material words, or codes.
- `system` (string, required): Which list to search.

Output parameters:

- `limit` (string): What it does not establish.
- `source` (object)
- `summary` (string): The answer, figure or code first.

### `site_tools` (~99 tokens)

Find a website tool for a file or record

List the ComplyOnSite website tools for work on a file or kept record: safety data sheet to Excel, Receipt of Waste workbook check, Defra report from a weighbridge export, RAMS review, paper form to phone form, site diary, issues, progress report and register. Returns each tool’s link and what it needs. This server does not read files.

Input parameters:

- `query` (string): Words for the job; omit for every tool.

Output parameters:

- `limit` (string): What it does not establish.
- `source` (object)
- `summary` (string): The answer, figure or code first.

## Diagnostics

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

## Score history

- 2026-11-05: 82
- 2026-11-04: 82
- 2026-11-03: 82
- 2026-11-02: 81
- 2026-11-01: 81
- 2026-10-31: 82
- 2026-10-30: 82
- 2026-10-29: 81
- 2026-10-28: 81
- 2026-10-27: 80
- 2026-10-26: 80
- 2026-10-25: 79
- 2026-10-24: 79
- 2026-10-23: 78
- 2026-10-22: 78
- 2026-10-21: 77
- 2026-10-20: 77
- 2026-10-19: 76
- 2026-10-18: 76
- 2026-10-17: 75
- 2026-10-16: 75
- 2026-10-15: 75
- 2026-10-14: 74
- 2026-10-13: 74
- 2026-10-12: 73
- 2026-10-10: 72

## Common questions

### What is the ComplyOnSite MCP server?

ComplyOnSite is an MCP server listed in the public MCP registry as com.complyonsite/tools. Waste codes, transfer notes filled in from chat, HAVS, noise and first-aid calculators. This page covers its hosted endpoint (https://mcp.complyonsite.com/mcp).

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

ComplyOnSite scores 82 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 ComplyOnSite MCP server expose?

ComplyOnSite exposes 11 tools: ewc_search, first_aid_requirements, havs_daily_exposure, noise_daily_exposure, construction_injury_statistics, and 6 more. Their descriptions and schemas cost roughly 1,865 tokens of context every time the server is loaded.

### Does the ComplyOnSite MCP server require authentication?

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

### Is the ComplyOnSite MCP server still maintained?

ComplyOnSite is still listed as active in the MCP registry. We last reached this channel on 5 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://mcp.complyonsite.com/mcp
- Website: https://complyonsite.com/docs/mcp
- Changelog RSS feed: https://verifymcp.io/servers/com-complyonsite-tools/mcp.xml
- Changelog JSON feed: https://verifymcp.io/servers/com-complyonsite-tools/mcp.json
- HTML version of this page: https://verifymcp.io/servers/com-complyonsite-tools/mcp