peaq Robotics

AdvisoryAudited by Static analysis on Apr 30, 2026.

Overview

No suspicious patterns detected.

Findings (0)

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

What this means

If invoked, the agent can make persistent identity, storage, or access-control changes in the configured peaq/ROS environment.

Why it was flagged

The skill exposes commands that can create/read identity data, write/read storage, and modify access roles or permissions through ROS 2 services.

Skill content
- DID create/read
- Storage add/read
- Access role/permission operations
Recommendation

Use the mutating commands only for explicit user requests, verify the target ROS domain and config, and review access-control changes before applying them.

What this means

Commands will act in the context of the local peaq robot configuration and any account or wallet identity that the ROS node uses.

Why it was flagged

The ROS node is started with a local peaq robot config, and the skill performs identity/access operations through that configured environment.

Skill content
Config YAML exists (either:
  - set `PEAQ_ROS2_CONFIG_YAML`, or
  - use default `<PEAQ_ROS2_ROOT>/peaq_ros2_examples/config/peaq_robot.yaml`).
Recommendation

Set `PEAQ_ROS2_ROOT` and `PEAQ_ROS2_CONFIG_YAML` explicitly, verify they point to the intended account/config, and avoid exposing sensitive config files.

What this means

A compromised or unexpected local ROS workspace could affect what actually runs when the skill starts nodes.

Why it was flagged

The reviewed skill delegates core runtime behavior to local ROS packages and workspace code that are not installed or fully provenance-verified by this skill.

Skill content
`peaq-robotics-ros2` workspace is already present and built.
Recommendation

Use only a trusted peaq-robotics-ros2 checkout/build and keep the ROS overlay environment under user control.

What this means

Sensitive data could be stored unintentionally, and retrieved JSON should not be blindly treated as trusted instructions.

Why it was flagged

The skill can write and later read JSON data through the storage service, including data that might be reused as agent state.

Skill content
"Read my DID, store this JSON as `agent_state`, then read it back."
Recommendation

Avoid storing secrets, clearly label stored state, and treat data read back from storage as untrusted unless the user verifies it.

NoteHigh Confidence
ASI10: Rogue Agents
What this means

ROS 2 nodes may remain active until stopped, potentially continuing to interact with the configured robot/peaq environment.

Why it was flagged

Start commands launch ROS 2 nodes in the background and store PID/log files, so processes can continue after the initial command returns.

Skill content
nohup "$@" >"$LOG_DIR/$name.log" 2>&1 &
Recommendation

Use the provided stop commands when finished and check the PID/log directories if behavior is unexpected.