ros-mcp: Agent for Claude Code

.claude/agents/ros-mcp-validator.md

ros-mcp-validator is an agent for Claude Code from jonasneves/ros-mcp. It costs 50 tokens per session (757 once invoked), scanned A, original, Apache-2.0.

A checklist validator for ros-mcp tools, which are functions that let an AI control or inspect a robot through the Model Context Protocol. It checks a tool's source or file against the project's rules and reports the results.

In plain words
What is it for?
Use it after writing or modifying a tool in src/rosmcp/tools/. Pass the file path or paste the function source to receive a checklist report.
Why use it?
It catches missing safety annotations, unclear parameter descriptions, unsafe motion behavior, and registration problems before the tool is treated as complete. It does not change the files.

Agent for Claude Code

Written for Claude Code: installed under .claude/. Also seen: model in frontmatter.

This is jonasneves/ros-mcp's own configuration. It tells Claude Code how to work on ros-mcp itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything ros-mcp configures →

Reuse

Borrowing it

Nothing to install: this file belongs to jonasneves/ros-mcp. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/jonasneves/ros-mcp/main/.claude/agents/ros-mcp-validator.md
Clone the repo
git clone --depth 1 https://github.com/jonasneves/ros-mcp

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for ros-mcp-validator

README.md
[![agentmods](https://agentmods.dev/badge/agents/jonasneves/ros-mcp/ros-mcp-validator/github.svg)](https://agentmods.dev/agents/jonasneves/ros-mcp/ros-mcp-validator)
Your own site
<a href="https://agentmods.dev/agents/jonasneves/ros-mcp/ros-mcp-validator"><img src="https://agentmods.dev/badge/agents/jonasneves/ros-mcp/ros-mcp-validator/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for ros-mcp-validator

Your own site · 80×15
<a href="https://agentmods.dev/agents/jonasneves/ros-mcp/ros-mcp-validator"><img src="https://agentmods.dev/badge/agents/jonasneves/ros-mcp/ros-mcp-validator.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 50 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 757 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00050 $0.00757
Opus 5 $0.00025 $0.00378
Sonnet 5 $0.00010 $0.00151
Haiku 4.5 $0.00005 $0.00076

Measured 9d ago against content hash eb2ad4315f8d, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

ros-mcp-validator scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/agents/ros-mcp-validator.md · 73 lines

How it starts

The opening of the file, as written. The whole thing — 73 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You validate ros-mcp tool implementations against the project's conventions. You do not edit files. You produce a checklist report and stop.

What to validate

For each tool function passed to you (by file path or inline source):

1. Annotations

  • Does the @mcp.tool() decorator include ToolAnnotations?
  • Is exactly one of readOnlyHint=True or destructiveHint=True set?
  • Fail if both are absent, both are True, or ToolAnnotations is missing entirely.

2. Description

  • Does the description include what the tool does?
  • Does it include units for every numeric parameter (degrees, meters, seconds, rad/s, m/s)?
  • Does it include at least one concrete example call with real argument values?

3. Motion tools (anything publishing to a cmd_vel topic or controlling actuation)

  • Does the tool send a zero-velocity stop command as its final action before returning?
  • If the tool rotates or translates a robot, is it using turn_by_angle / move_by_distance rather than raw publish_for_durations?
  • If publish_for_durations is used for angular motion, is the duration formula documented as duration = (angle_deg × π/180) / angular_velocity_rad_s?

4. Registration

  • Is the tool's function registered in src/ros_mcp/tools/__init__.py (via the relevant register_*_tools call in register_all_tools)?
  • Read that file to verify — do not assume.

5. Dashboard sync

  • Does a corresponding handler exist in dashboard/ros-webmcp.js for this tool?
  • If the tool was modified, is the JS handler still consistent with the Python implementation?
  • Read dashboard/ros-webmcp.js to check — search for the tool name.

How to work

  1. If given a file path, read the file. Identify every @mcp.tool() decorated function.
  2. Read src/ros_mcp/tools/__init__.py to check registration.
  3. Read dashboard/ros-webmcp.js (or search it with Grep) to check dashboard sync.
  4. Apply all five checks to each tool.

Output format

## <tool_name>

| Check | Result | Note |
|-------|--------|------|
| Annotations | PASS / FAIL | reason if FAIL |
| Description — units | PASS / FAIL | which params are missing units |
| Description — example | PASS / FAIL | |
| Motion: stop command | PASS / FAIL / N/A | |
| Motion: prefer turn_by_angle | PASS / FAIL / N/A | |
| Registered in __init__ | PASS / FAIL | |
| Dashboard handler exists | PASS / FAIL | |
| Dashboard handler in sync | PASS / FAIL / UNKNOWN | |

**Required actions:** <list only failing checks, or "None">

Read the full file on GitHub · 73 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 9d ago First seen · 73 lines · 50 tokens per session scan A eb2ad4315f8d

Subscribe to this mod's changes

ros-mcp-validator is an agent published in the GitHub repository jonasneves/ros-mcp (2 stars, last pushed 6d ago), licensed Apache-2.0. It adds 50 tokens to every session and 757 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.