mujoco-mcp

Overview Schema Source Related Servers Discussions Admin

mujoco_step_guarded

Advance a paused MuJoCo session step-by-step, monitoring contact forces and stopping when a specified force limit is exceeded. Returns peak force and world-frame resultants for safety validation.

Instructions

Advance a paused session, measuring all contacts after every physics step.

Stops at the first force-limit exceedance. guard.peak_force_N is the maximum
summed contact load on body_name (including descendants); opposite contact
forces do not cancel. World-frame resultant force and torque are also returned.
The guard detects an exceedance, it cannot guarantee zero overshoot. No hard
real-time or physical safety certification is implied. A tripped session is
latched until reset or snapshot restore. Optional record_every samples a JSON
trace every N steps (0 disables; maximum 2000 frames). Initial, terminal and
trip states are always captured; peak_force_N tracks all steps, not samples.
Persist returned trace/provenance in the consuming project for replay/evidence.

Input Schema

TableJSON Schema

Name	Required	Description
`ctrl`	No	Actuator controls: full vector in actuator order, or {actuator_name: value} for a subset. Persists on the session until changed.
`nsteps`	No
`body_name`	Yes
`session_id`	Yes
`force_limit`	Yes
`record_every`	No

Output Schema

TableJSON Schema

Name	Required	Description	Default
`result`	Yes

Schema Changelog

Changes observed during successful MCP inspections.

First observedv0.1.0Sep 20, 2026

TDQS

A4.4/5.0

Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

The description discloses critical behavioral nuances beyond the annotations: the guard cannot guarantee zero overshoot, no safety certification, latching until reset/restore, sampling behavior (record_every, max 2000 frames), and that peak_force_N tracks all steps not just samples. These details are not in the annotations and are essential for correct usage.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is dense but well-structured, front-loading the core action and guard behavior. Each sentence adds unique value (overshoot, latching, sampling, persistence). It is longer than minimal but not verbose; the structure keeps the most important behavior first.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given the tool's complexity (force guard, sampling, latching) and the presence of an output schema, the description covers all essential usage aspects: when it stops, what peak_force_N means, trip latching, sampling details, and persistence guidance. Nothing an agent needs to invoke it correctly is missing.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

With schema description coverage at only 17%, the description compensates by explaining the meaning of record_every (sampling trace, max 2000 frames), ctrl persistence, and the semantics of peak_force_N (summed contact load, no cancellation). However, it does not elaborate on nsteps or force_limit beyond what titles imply, and there is a minor discrepancy (schema max 10000 for record_every vs description's 2000) that could confuse.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states the tool's purpose: 'Advance a paused session, measuring all contacts after every physics step' and specifies the stopping condition ('Stops at the first force-limit exceedance'). It distinguishes itself from the sibling mujoco_step by adding the force guard, making the verb+resource+differentiator explicit.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines3/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description provides context (advancing a paused session, guard behavior) but does not explicitly compare with alternatives like mujoco_step or state when the unguarded version is preferable. The existence of a sibling tool is implied but not referenced, so an agent must infer the differentiation.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Deploy Server