# armguard-mcp (pypi · armguard-mcp)

Safety-first MCP server for ROS 2 robot arms: server-side safety envelope, human approval, audit

- Trust score: 80/100 (high trust)
- Change this week: 0
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-11-05

## Components

- pypi · `armguard-mcp`: 80/100 (this document), [markdown](https://verifymcp.io/servers/lynn-hh-armguard-mcp/armguard-mcp.md), [page](https://verifymcp.io/servers/lynn-hh-armguard-mcp/armguard-mcp)

## Channel facts

- Registry: `pypi`
- Package: `armguard-mcp`
- Version: `0.1.1`
- 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 2026-11-05.

- **Supply Chain Security**: 100/100
  - No malware found by supply-chain analysis.
  - 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.
  - 1 of 29 dependencies flagged as unhealthy.
- **Provenance & Transparency**: 97/100
  - Source repository is publicly reachable at the declared URL.
  - Cryptographically verified build provenance (signed, bound to Lynn-hh/armguard-mcp).
  - Clear OSI-approved license (Apache-2.0).
  - Actively maintained (last published 11 days ago).
  - Disclosure check failed: no security disclosure policy was found in the source repository.
- **Schema Quality & AI Usability**: 73/100
  - AI-judged instruction clarity (good).
  - Tool/resource definitions use about 1800 tokens (~81/item across 22 items; 22 tools + 0 resources), lean.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 0/100
  - Stability not yet verified: not enough scan history yet (needs a 30-day window).
- **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 (95% 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.
  - All 1 tool(s) whose name or description implies an irreversible operation declare an MCP destructiveHint annotation.
  - An AI judge read all 23 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).

**Unverified: 1 category.** A category scored 0 because we could not verify it: a data source with nothing on this package, evidence we could not reach, or a check we could not run. We only credit what we can confirm.

## Install

### How do I install the armguard-mcp server?

armguard-mcp runs locally as a PyPI package, launched with uvx armguard-mcp --backend fake. Ready-made configuration for Claude, Cursor, VS Code, Codex and 3 more is on this page, copied from each client's own documentation.

### Claude

```bash
claude mcp add lynn-hh-armguard-mcp -- uvx armguard-mcp --backend fake
```

### Cursor

```json
{
  "mcpServers": {
    "lynn-hh-armguard-mcp": {
      "command": "uvx",
      "args": [
        "armguard-mcp",
        "--backend",
        "fake"
      ]
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "lynn-hh-armguard-mcp": {
      "command": "uvx",
      "args": [
        "armguard-mcp",
        "--backend",
        "fake"
      ]
    }
  }
}
```

### Codex

```bash
codex mcp add lynn-hh-armguard-mcp -- uvx armguard-mcp --backend fake
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "lynn-hh-armguard-mcp": {
      "type": "local",
      "command": [
        "uvx",
        "armguard-mcp",
        "--backend",
        "fake"
      ],
      "enabled": true
    }
  }
}
```

### Hermes

```yaml
mcp_servers:
  lynn-hh-armguard-mcp:
    command: "uvx"
    args: ["armguard-mcp", "--backend", "fake"]
```

### Netclaw

```json
{
  "McpServers": {
    "lynn-hh-armguard-mcp": {
      "Transport": "stdio",
      "Command": "uvx",
      "Arguments": [
        "armguard-mcp",
        "--backend",
        "fake"
      ]
    }
  }
}
```

### Other

```json
{
  "mcpServers": {
    "lynn-hh-armguard-mcp": {
      "command": "uvx",
      "args": [
        "armguard-mcp",
        "--backend",
        "fake"
      ]
    }
  }
}
```

## 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.

### 2026-10-30 (score 80, 0)

- [functional regression] Security disclosure: unverified → fail

### 2026-10-29 (score 80, 0)

- [functional regression] Security disclosure: fail → unverified

### 2026-10-27 (score 80, 0)

- [functional] Package version: 0.1.0 → 0.1.1

### 2026-10-26 (score 80, +15)

- [security improvement] Malware scan: unverified → pass

### 2026-10-25 (score 65, +16)

- [security regression] Malware scan: pass → unverified
- [security improvement] Injection markers: unverified → pass
- [security] First check of Judged manipulation: pass
- [security] Stability: Stability not yet verified: not enough scan history yet (needs a 30-day window).
- [functional improvement] Tool coverage: unverified → 100
- [functional improvement] MCP protocol: unverified → pass
- [functional] First check of Schema quality: fail
- [functional] First check of Destructive annotations: 100
- [functional] First check of Schema quality: pass
- [functional] First check of Schema quality: good
- [functional] First check of Tool coverage: 86
- [functional] First check of Tool coverage: 95
- [functional] Package version: 0.1.0 → 0.1.1

### 2026-10-20 (score 49, +15)

- [security improvement] Malware scan: unverified → pass

### 2026-10-19 (score 34)

First indexed and scored.

## MCP tools (22)

### `get_robot_state` (~50 tokens)

Current joint positions [rad], TCP pose [m, quaternion xyzw] in the robot base frame, estimated
external wrench [N, N*m], gripper state and safety status. Read-only.

Output parameters:

- `backend` (string)
- `ee_pose`
- `gripper`
- `joint_state`
- `robot` (string)
- `safety`
- `wrench`

### `get_safety_envelope` (~63 tokens)

The limits the server enforces (joint limits, workspace box, keep-out zones, speed/step caps,
force thresholds, gripper limits, approval mode). Read this before planning and stay inside it:
violations are rejected server-side and cannot be overridden.

Output parameters:

- `approval`
- `base_frame` (string)
- `camera_topics` (array)
- `controllers_allowlist` (array)
- `dry_run` (boolean)
- `ee_frame` (string)
- `enabled_tool_groups` (array)
- `force`
- `gripper`
- `home_joint_positions` (array)
- `joint_limits` (object)
- `joint_names` (array)
- `max_image_width` (integer)
- `motion`
- `rate_limits`
- `robot` (string)
- `summary` (string)
- `workspace`

### `lookup_transform` (~73 tokens)

Pose of source_frame expressed in target_frame (tf2 semantics), e.g. target_frame='fr3_link0',
source_frame='fr3_hand_tcp'. Position in metres, orientation as a quaternion (x, y, z, w).

Input parameters:

- `source_frame` (string, required)
- `target_frame` (string, required)

Output parameters:

- `frame_id` (string): Frame the pose is expressed in
- `orientation`: Unit quaternion (x, y, z, w)
- `position`

### `list_controllers` (~28 tokens)

ros2_control controllers with their type and state, plus which ones the policy lets you switch.

Output parameters:

- `controllers` (array)
- `switchable` (array): Controllers the policy allows to be (de)activated

### `list_ros_graph` (~39 tokens)

ROS 2 nodes, topics, services and actions visible to the server. Read-only; you cannot publish
or call arbitrary services through this server.

Output parameters:

- `actions` (array)
- `nodes` (array)
- `services` (array)
- `topics` (array)

### `get_audit_tail` (~41 tokens)

The last n audit-log events (tool calls, approvals, denials, e-stops). Images are redacted.

Input parameters:

- `n` (integer)

Output parameters:

- `events` (array)

### `camera_snapshot` (~70 tokens)

Grab the latest image from an allowlisted camera topic (see get_safety_envelope.camera_topics).
Returns the image plus JSON metadata (width, height, stamp). Images wider than the policy's
max_image_width are downscaled (or refused if Pillow is not installed).

Input parameters:

- `topic` (string, required)

### `plan_to_joints` (~159 tokens)

Plan (do NOT move) a joint-space motion from the current state to joint_positions [rad].
Returns a plan summary validated against the safety envelope; if status is not 'rejected',
pass plan_id to execute_plan. Scaling above the policy cap is clamped.

Input parameters:

- `acceleration_scaling`: Fraction of max joint accel (0, 1]
- `joint_positions` (array, required): Target joint positions [rad], one per joint in the order ['fr3_joint1', 'fr3_joint2', 'fr3_joint3', 'fr3_joint4', 'fr3_joint5', 'fr3_joint6', 'fr3_joint7']
- `velocity_scaling`: Fraction of max joint speed (0, 1]

Output parameters:

- `acceleration_scaling` (number)
- `dry_run` (boolean)
- `duration_s` (number)
- `executable` (boolean)
- `expires_at`: ISO-8601 UTC expiry of the plan handle
- `final_ee_pose`
- `final_joint_positions` (array)
- `force_monitoring` (string): Whether the backend currently provides an external wrench estimate. Without one the server cannot enforce the contact force limit, and execute_plan refuses to move unless the policy sets force.requir…
- `kind` (string)
- `max_joint_acceleration_ratio` (number): Peak joint acceleration (estimated from the waypoint timing) as a fraction of the joint limit
- `max_joint_travel_rad` (number)
- `max_joint_velocity_ratio` (number): Peak joint speed as a fraction of the joint limit
- `notes` (array)
- `num_waypoints` (integer)
- `plan_id` (string)
- `requires_approval` (boolean)
- `start_joint_positions` (array)
- `status` (string): rejected plans can never be executed; needs_approval requires a human to approve
- `tcp_path_length_m` (number)
- `velocity_scaling` (number)
- `violations` (array)

### `plan_to_pose` (~149 tokens)

Plan (do NOT move) a motion that brings the TCP to a pose. Uses inverse kinematics; the path is
a joint-space interpolation (the TCP does not move in a straight line - use plan_cartesian_path
for that). Returns a plan summary validated against the safety envelope.

Input parameters:

- `acceleration_scaling`: Fraction of max joint accel (0, 1]
- `frame_id`: Frame of the target; default = robot base frame
- `orientation`: Target TCP orientation (x, y, z, w); omit to keep current
- `position` (required): Target TCP position [m]
- `velocity_scaling`: Fraction of max joint speed (0, 1]

Output parameters:

- `acceleration_scaling` (number)
- `dry_run` (boolean)
- `duration_s` (number)
- `executable` (boolean)
- `expires_at`: ISO-8601 UTC expiry of the plan handle
- `final_ee_pose`
- `final_joint_positions` (array)
- `force_monitoring` (string): Whether the backend currently provides an external wrench estimate. Without one the server cannot enforce the contact force limit, and execute_plan refuses to move unless the policy sets force.requir…
- `kind` (string)
- `max_joint_acceleration_ratio` (number): Peak joint acceleration (estimated from the waypoint timing) as a fraction of the joint limit
- `max_joint_travel_rad` (number)
- `max_joint_velocity_ratio` (number): Peak joint speed as a fraction of the joint limit
- `notes` (array)
- `num_waypoints` (integer)
- `plan_id` (string)
- `requires_approval` (boolean)
- `start_joint_positions` (array)
- `status` (string): rejected plans can never be executed; needs_approval requires a human to approve
- `tcp_path_length_m` (number)
- `velocity_scaling` (number)
- `violations` (array)

### `plan_cartesian_path` (~119 tokens)

Plan (do NOT move) a straight-line TCP path through the waypoints [m]. Total TCP path length is
capped by the policy (max_cartesian_step_m). Returns a plan summary validated against the envelope.

Input parameters:

- `acceleration_scaling`: Fraction of max joint accel (0, 1]
- `frame_id`: Frame of the waypoints; default = base frame
- `velocity_scaling`: Fraction of max joint speed (0, 1]
- `waypoints` (array, required): TCP waypoints, in order

Output parameters:

- `acceleration_scaling` (number)
- `dry_run` (boolean)
- `duration_s` (number)
- `executable` (boolean)
- `expires_at`: ISO-8601 UTC expiry of the plan handle
- `final_ee_pose`
- `final_joint_positions` (array)
- `force_monitoring` (string): Whether the backend currently provides an external wrench estimate. Without one the server cannot enforce the contact force limit, and execute_plan refuses to move unless the policy sets force.requir…
- `kind` (string)
- `max_joint_acceleration_ratio` (number): Peak joint acceleration (estimated from the waypoint timing) as a fraction of the joint limit
- `max_joint_travel_rad` (number)
- `max_joint_velocity_ratio` (number): Peak joint speed as a fraction of the joint limit
- `notes` (array)
- `num_waypoints` (integer)
- `plan_id` (string)
- `requires_approval` (boolean)
- `start_joint_positions` (array)
- `status` (string): rejected plans can never be executed; needs_approval requires a human to approve
- `tcp_path_length_m` (number)
- `velocity_scaling` (number)
- `violations` (array)

### `execute_plan` (~136 tokens)

EXECUTE a previously planned motion on the robot. The server re-validates the plan against the
safety envelope, rejects stale/expired/used plans, may ask a human to approve (depending on
policy), monitors contact force while moving (aborting and latching an e-stop above the limit),
and streams progress. Plan handles are single-use and one plan executes at a time. Refused if the
robot provides no force estimate (unless the policy allows it). Cancelling the call stops the
robot. In dry-run mode nothing moves.

Input parameters:

- `plan_id` (string, required): plan_id returned by a plan_* tool

Output parameters:

- `approval` (string): How the execution was authorised
- `dry_run` (boolean)
- `duration_s` (number)
- `executed` (boolean)
- `final_ee_pose`
- `final_joint_positions`
- `max_observed_force_n`
- `message` (string)
- `plan_id` (string)
- `status` (string)

### `get_motion_status` (~30 tokens)

Whether a plan is executing, its progress (0..1) and the result of the last execution.

Output parameters:

- `executing` (boolean)
- `last_result`
- `message` (string)
- `plan_id`
- `progress` (number): 0..1 fraction of the active plan

### `gripper_move` (~61 tokens)

Move the gripper fingers to an opening width [m] without applying grasp force.

Input parameters:

- `speed_mps`: Finger speed [m/s]; default = policy max
- `width_m` (number, required): Target finger opening [m], within the policy range

Output parameters:

- `data` (object)
- `message` (string)
- `ok` (boolean)

### `gripper_grasp` (~168 tokens)

Close the gripper on an object of about width_m [m] with force_n [N]. Succeeds (ok=true) only
if the fingers stop within [width - epsilon_inner, width + epsilon_outer]. Forces above the
policy maximum are rejected, not clamped.

Input parameters:

- `epsilon_inner_m` (number): Tolerance [m]: the grasp succeeds down to width_m - epsilon_inner_m
- `epsilon_outer_m` (number): Tolerance [m]: the grasp succeeds up to width_m + epsilon_outer_m
- `force_n` (number, required): Grasp force [N]; must not exceed the policy maximum
- `speed_mps`: Finger speed [m/s]; default = policy max
- `width_m` (number, required): Expected object width [m]

Output parameters:

- `data` (object)
- `message` (string)
- `ok` (boolean)

### `gripper_home` (~21 tokens)

Home (fully open and calibrate) the gripper.

Output parameters:

- `data` (object)
- `message` (string)
- `ok` (boolean)

### `switch_controllers` (~67 tokens)

Activate/deactivate ros2_control controllers. Only controllers in the policy allowlist may be
switched; a human may need to approve.

Input parameters:

- `activate` (array): Controllers to activate (must be allowlisted)
- `deactivate` (array): Controllers to deactivate (must be allowlisted)

Output parameters:

- `controllers` (array)
- `switchable` (array): Controllers the policy allows to be (de)activated

### `set_collision_thresholds` (~71 tokens)

Set the robot's collision (reflex) thresholds. Values above the policy maxima are rejected.

Input parameters:

- `force_n` (number, required): Collision force threshold [N]; <= policy max_contact_force_n
- `torque_nm` (number, required): Collision torque threshold [N*m]; <= policy max_contact_torque_nm

Output parameters:

- `data` (object)
- `message` (string)
- `ok` (boolean)

### `stop_motion` (~50 tokens)

Stop the current motion immediately: the arm trajectory and any gripper action. Always
available, never rate limited, never needs approval. Does not latch; use estop to also block
further motion.

Output parameters:

- `data` (object)
- `message` (string)
- `ok` (boolean)

### `estop` (~87 tokens)

SOFTWARE E-STOP: stop all motion (arm and gripper), invalidate every plan and refuse
motion/gripper/control tools until a human approves reset_estop. Always available, never rate
limited, never needs approval, accepts any reason. This is not a substitute for the hardware
e-stop.

Input parameters:

- `reason`: Optional short reason, shown to the human who resets the e-stop

Output parameters:

- `dry_run` (boolean)
- `estop_event`: Number of the latched e-stop event
- `estopped` (boolean)
- `estopped_at`
- `executing_plan_id`
- `force_violation_latched` (boolean)
- `last_violation`
- `note` (string)
- `reason`
- `reason_source`: Who gave the reason: the AI agent (via estop) or the server itself

### `reset_estop` (~71 tokens)

Release the software e-stop (and any latched force violation). ALWAYS requires human approval via
the MCP client, regardless of the approval mode. An approval only releases the e-stop event it was
shown for; if another e-stop is latched while the prompt is open, call reset_estop again.

Output parameters:

- `dry_run` (boolean)
- `estop_event`: Number of the latched e-stop event
- `estopped` (boolean)
- `estopped_at`
- `executing_plan_id`
- `force_violation_latched` (boolean)
- `last_violation`
- `note` (string)
- `reason`
- `reason_source`: Who gave the reason: the AI agent (via estop) or the server itself

### `error_recovery` (~45 tokens)

Clear the robot's error/reflex state (e.g. after a collision reflex). May require human approval.
Refused while the software e-stop is active (reset it first).

Output parameters:

- `data` (object)
- `message` (string)
- `ok` (boolean)

### `get_safety_status` (~32 tokens)

E-stop state, latched force violation, dry-run flag and the last envelope violation. Never rate limited.

Output parameters:

- `dry_run` (boolean)
- `estop_event`: Number of the latched e-stop event
- `estopped` (boolean)
- `estopped_at`
- `executing_plan_id`
- `force_violation_latched` (boolean)
- `last_violation`
- `note` (string)
- `reason`
- `reason_source`: Who gave the reason: the AI agent (via estop) or the server itself

## Diagnostics

Captured diagnostic sections: Provenance, Install scripts, Dependencies. The full working is on the page: https://verifymcp.io/servers/lynn-hh-armguard-mcp/armguard-mcp#diagnostics

## Score history

- 2026-11-05: 80
- 2026-11-04: 80
- 2026-11-03: 80
- 2026-11-02: 80
- 2026-11-01: 80
- 2026-10-31: 80
- 2026-10-30: 80
- 2026-10-29: 80
- 2026-10-28: 80
- 2026-10-27: 80
- 2026-10-26: 80
- 2026-10-25: 65
- 2026-10-24: 49
- 2026-10-23: 49
- 2026-10-22: 49
- 2026-10-21: 49
- 2026-10-20: 49
- 2026-10-19: 34

## Common questions

### What is the armguard-mcp server?

armguard-mcp is listed in the public MCP registry as io.github.Lynn-hh/armguard-mcp. Safety-first MCP server for ROS 2 robot arms: server-side safety envelope, human approval, audit. This page covers its PyPI package (armguard-mcp).

### Is the armguard-mcp server safe to use?

armguard-mcp scores 80 out of 100 on VerifyMCP. We found no known CVEs affecting it as of 5 November 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 armguard-mcp server expose?

armguard-mcp exposes 22 tools: get_robot_state, get_safety_envelope, lookup_transform, list_controllers, list_ros_graph, and 17 more. Their descriptions and schemas cost roughly 1,630 tokens of context every time the server is loaded.

### Is the armguard-mcp server still maintained?

armguard-mcp 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.

### What licence is the armguard-mcp server under?

armguard-mcp 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.

### What did VerifyMCP test on the armguard-mcp server?

VerifyMCP ran 15 checks on armguard-mcp, covering Malware scan, Known CVEs, Install scripts, Dependency health and 11 more. One item could not be completed, a gap on our side rather than a finding about armguard-mcp: Stability. One category scores 0 for that reason, because we only credit what we can confirm. Endpoint Security and Transport & Reachability do not apply to a PyPI package, so they were not tested. The score does not measure everything, and the scoring documentation lists what it leaves out.

## Links

- PyPI project: https://pypi.org/project/armguard-mcp/
- Socket report: https://socket.dev/pypi/package/armguard-mcp
- Repository: https://github.com/Lynn-hh/armguard-mcp
- Changelog RSS feed: https://verifymcp.io/servers/lynn-hh-armguard-mcp/armguard-mcp.xml
- Changelog JSON feed: https://verifymcp.io/servers/lynn-hh-armguard-mcp/armguard-mcp/changelog.json
- Trust report (JSON): https://verifymcp.io/servers/lynn-hh-armguard-mcp/armguard-mcp.json
- HTML version of this page: https://verifymcp.io/servers/lynn-hh-armguard-mcp/armguard-mcp