# EPICS Channel Access (pypi · labmcp-epics)

MCP server for EPICS Channel Access: read, monitor and guarded writes of process variables.

- Trust score: 68/100 (medium)
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-09-30

## Components

- pypi · `labmcp-epics`: 68/100 (this document), [markdown](https://verifymcp.io/servers/k-dense-ai-labmcp-epics/labmcp-epics.md), [page](https://verifymcp.io/servers/k-dense-ai-labmcp-epics/labmcp-epics)

## Channel facts

- Registry: `pypi`
- Package: `labmcp-epics`
- Version: `0.1.2`
- Transport: `stdio`

## Trust breakdown

How this component scores in each security and reliability category. Every signal is checked automatically from public evidence about the published package, including repeated runs of it in an isolated sandbox, 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 2026-09-30.

- **Supply Chain Security**: 49/100
  - Malware scan not yet available for this package.
  - No known CVEs affecting this package version or its production dependencies.
  - Runs hatchling.build at install time, a recognised build step with no custom scripting around it.
  - 3 of 19 dependencies flagged as unhealthy.
- **Provenance & Transparency**: 100/100
  - Source repository is publicly reachable at the declared URL.
  - Cryptographically verified build provenance (signed, bound to K-Dense-AI/lab-instrument-mcps).
  - Clear OSI-approved license (Apache-2.0).
  - Actively maintained (last published 3 days ago).
  - Publishes a security disclosure policy (SECURITY.md).
- **Schema Quality & AI Usability**: 76/100
  - AI-judged instruction clarity (excellent).
  - Context-footprint check failed: tool/resource definitions use about 1287 tokens (~128/item across 10 items; 10 tools + 0 resources), over budget; trim descriptions and params.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 13/100
  - Stability observed for 4 of 30 days with no destabilising changes; credit accrues until the full window elapses.
- **Tool Coverage**: 98/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 94% 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 10 captured tool definition(s), and no name or description among them implies an irreversible operation.
  - An AI judge read all 11 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 EPICS Channel Access MCP server?

EPICS Channel Access runs locally as a PyPI package, launched with uvx labmcp-epics. 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 k-dense-ai-labmcp-epics -- uvx labmcp-epics
```

### Cursor

```json
{
  "mcpServers": {
    "k-dense-ai-labmcp-epics": {
      "command": "uvx",
      "args": [
        "labmcp-epics"
      ]
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "k-dense-ai-labmcp-epics": {
      "command": "uvx",
      "args": [
        "labmcp-epics"
      ]
    }
  }
}
```

### Codex

```bash
codex mcp add k-dense-ai-labmcp-epics -- uvx labmcp-epics
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "k-dense-ai-labmcp-epics": {
      "type": "local",
      "command": [
        "uvx",
        "labmcp-epics"
      ],
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add k-dense-ai-labmcp-epics --command uvx --arg labmcp-epics
```

### Hermes

```yaml
mcp_servers:
  k-dense-ai-labmcp-epics:
    command: "uvx"
    args: ["labmcp-epics"]
```

### Netclaw

```json
{
  "McpServers": {
    "k-dense-ai-labmcp-epics": {
      "Transport": "stdio",
      "Command": "uvx",
      "Arguments": [
        "labmcp-epics"
      ]
    }
  }
}
```

### Vellum

```bash
assistant mcp add k-dense-ai-labmcp-epics -t stdio -c uvx -a labmcp-epics
```

### Other

```json
{
  "mcpServers": {
    "k-dense-ai-labmcp-epics": {
      "command": "uvx",
      "args": [
        "labmcp-epics"
      ]
    }
  }
}
```

## 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.

### 2026-09-29 (score 68, +1)

No change was recorded against any check on this day. Stability & Change Management went from 7 to 10. That category is still filling its 30-day observation window: 2 days of observed history at the previous scan, 3 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-09-28 (score 67, +28)

- [functional] We updated how we score, so this day's move reflects our rubric, not a change to the server

### 2026-09-27 (score 39, −32)

- [security regression] Tool safety: pass → unverified
- [security] Stability: Stability not yet verified: we do not have a sandbox capture of the MCP schema this version of the package serves yet.
- [functional regression] Capabilities: pass → unverified
- [functional regression] Tool coverage: 100 → unverified
- [functional] First check of Schema quality: unverified
- [functional] Package version: 0.1.1 → 0.1.2

### 2026-09-26 (score 71)

First indexed and scored.

## MCP tools (10)

### `get_connection_info` (~40 tokens)

Get Connection Info

Report which instrument is connected (identity, address, simulated or real),
whether the server is read-only, and the active safety limits. Call this first.

### `get_command_log` (~46 tokens)

Get Command Log

Return the most recent raw commands sent to / replies received from the
instrument (newest last). Useful for debugging and for recording what was done.

Input parameters:

- `limit` (integer)

Output parameters:

- `result` (array)

### `reconnect` (~35 tokens)

Reconnect

Close and re-open the connection to the instrument (e.g. after it was power
cycled or a cable was re-plugged).

### `get_pv` (~96 tokens)

Get Pv

Read one PV with its metadata: value, units, precision, alarm severity/status, IOC timestamp
and age, display/alarm/warning/control limits and enum state names. Waveforms come back
downsampled with min/max/mean statistics.

Input parameters:

- `max_elements` (integer): Arrays longer than this are downsampled
- `name` (string, required): PV name, e.g. 'BL7:MTR1.RBV'

Output parameters:

- `age_s`: Seconds since the IOC timestamp (large = stale value)
- `alarm_limits`: LOLO/HIHI
- `array_summary`
- `as_string`: CHAR waveform decoded as a (long) string
- `control_limits`: DRVL/DRVH for output records; writes must lie inside
- `display_limits`
- `downsampled` (boolean)
- `element_count` (integer)
- `enum_strings`
- `name` (string)
- `native_type` (string): CA native type: DOUBLE, FLOAT, LONG, INT, CHAR, ENUM or STRING
- `precision`
- `severity` (string): Alarm severity: NO_ALARM, MINOR_ALARM, MAJOR_ALARM or INVALID_ALARM
- `status` (string): Alarm status, e.g. NO_ALARM, HIHI, HIGH, LOW, LOLO, UDF, COMM, TIMEOUT
- `timestamp`: IOC timestamp of the value (UTC)
- `units`
- `value`: Scalar, string, enum state name, or (possibly downsampled) array
- `warning_limits`: LOW/HIGH

### `get_pvs` (~75 tokens)

Get Pvs

Read many PVs at once (e.g. all motors or vacuum gauges of a beamline). PVs that cannot be
reached are listed in `errors` instead of failing the whole call.

Input parameters:

- `max_elements` (integer): Arrays longer than this are downsampled
- `names` (array, required): PV names

Output parameters:

- `errors` (array)
- `readings` (array)
- `timestamp` (string)

### `pv_info` (~69 tokens)

Pv Info

Connection details of a PV: serving IOC host:port, native type, element count, and whether this
client has read/write access (EPICS access security) and passes the put allow-list.

Input parameters:

- `name` (string, required): PV name, e.g. 'BL7:MTR1.RBV'

Output parameters:

- `connected` (boolean)
- `element_count` (integer)
- `name` (string)
- `native_type` (string)
- `put_allowed_by_allowlist` (boolean)
- `read_access` (boolean)
- `server` (string): host:port of the IOC (or CA gateway) serving the PV
- `timestamp` (string)
- `write_access` (boolean): EPICS access security grants this client write access

### `monitor_pv` (~116 tokens)

Monitor Pv

Subscribe to a PV and collect every value change for `duration_s` seconds (or until
\`max_updates`), then return statistics (min/max/mean/std, drift rate) and the updates.

Input parameters:

- `duration_s` (number): How long to collect updates
- `max_points` (integer): Updates to include in the reply (evenly thinned)
- `max_updates` (integer): Stop after this many updates
- `name` (string, required): PV name, e.g. 'BL7:MTR1.RBV'

Output parameters:

- `duration_s` (number)
- `n_updates` (integer)
- `name` (string)
- `non_finite_updates` (integer): Numeric updates that were NaN or infinite (not in stats)
- `severities_seen` (array)
- `stats`: For numeric PVs (arrays: per-update mean), finite values only
- `timestamp` (string)
- `updates` (array): Updates (arrays summarised by their mean), at most max_points

### `put_pv` (~171 tokens)

Put Pv

Write a PV. This can move motors, open shutters, change magnet or high-voltage setpoints and
heat or cool samples: tell the user exactly what will change first. The write is refused unless
the PV matches the put allow-list, the IOC grants write access, and the value has the right
type and lies within the PV's control limits (DRVL/DRVH). Returns the read-back value.

Input parameters:

- `name` (string, required): PV name, e.g. 'BL7:MTR1.RBV'
- `timeout_s` (number): Max seconds to wait for completion
- `value` (required): New value: number, enum state name or index, string, or array for waveforms
- `wait` (boolean): Wait for the IOC's put-completion (put-callback)

Output parameters:

- `completed` (boolean): True if the IOC confirmed completion (put-callback)
- `elapsed_s` (number)
- `name` (string)
- `previous_value`
- `requested`
- `severity` (string)
- `status` (string)
- `timestamp` (string)
- `value`: Value read back from the PV after the write
- `warnings` (array)

### `put_pvs` (~129 tokens)

Put Pvs

Write several PVs in order (e.g. set both slit blades). Every write is validated first (allow-
list, access rights, type, control limits); if any is invalid nothing is written. Writing stops
at the first failure and the rest are reported as not written. The whole batch must finish within
an hour: writes that would start later are not sent.

Input parameters:

- `timeout_s` (number): Max seconds to wait per write
- `wait` (boolean): Wait for each put-completion before the next write
- `writes` (array, required): PV writes, applied in order

Output parameters:

- `error`
- `not_written` (array): PVs not written because an earlier write failed
- `results` (array)
- `timestamp` (string)

### `apply_safe_state` (~68 tokens)

Apply Safe State

Emergency action: write the scientist-configured safe-state PVs (--option safe_state, e.g.
motor STOP fields, shutter close, HV off), all of them even if one fails. Available in read-only
mode. It does not replace the facility's own interlocks and stop buttons.

Output parameters:

- `configured` (boolean)
- `errors` (array)
- `message` (string)
- `results` (array)
- `timestamp` (string)

## Diagnostics

Captured diagnostic sections: Provenance, Install scripts, Dependencies. The full working is on the page: https://verifymcp.io/servers/k-dense-ai-labmcp-epics/labmcp-epics#diagnostics

## Score history

- 2026-09-30: 68
- 2026-09-29: 68
- 2026-09-28: 67
- 2026-09-27: 39
- 2026-09-26: 71

## Common questions

### What is the EPICS Channel Access MCP server?

EPICS Channel Access is an MCP server listed in the public MCP registry as io.github.K-Dense-AI/labmcp-epics. MCP server for EPICS Channel Access: read, monitor and guarded writes of process variables. This page covers its PyPI package (labmcp-epics).

### Is the EPICS Channel Access MCP server safe to use?

EPICS Channel Access scores 68 out of 100 on VerifyMCP. We found no known CVEs affecting it as of 30 September 2026. Its build provenance is signed and verified. 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 EPICS Channel Access MCP server expose?

EPICS Channel Access exposes 10 tools: get_connection_info, get_command_log, reconnect, get_pv, get_pvs, and 5 more. Their descriptions and schemas cost roughly 845 tokens of context every time the server is loaded.

### Is the EPICS Channel Access MCP server still maintained?

EPICS Channel Access is still listed as active in the MCP registry. We last reached this channel on 30 September 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.

### What licence is the EPICS Channel Access MCP server under?

EPICS Channel Access declares the Apache-2.0 licence, which is OSI-approved. That covers the source only, and says nothing about the cost of any service it calls.

## Links

- PyPI project: https://pypi.org/project/labmcp-epics/
- Socket report: https://socket.dev/pypi/package/labmcp-epics
- Repository: https://github.com/K-Dense-AI/lab-instrument-mcps
- Website: https://github.com/K-Dense-AI/lab-instrument-mcps/tree/main/servers/protocols/epics
- Changelog RSS feed: https://verifymcp.io/servers/k-dense-ai-labmcp-epics/labmcp-epics.xml
- Changelog JSON feed: https://verifymcp.io/servers/k-dense-ai-labmcp-epics/labmcp-epics.json
- HTML version of this page: https://verifymcp.io/servers/k-dense-ai-labmcp-epics/labmcp-epics