Skip to content
Back to the index

roboharm

PyPIInfrastructure

PulseGate's liveness check found it on 21 Sep 2026; it is registered on PyPI and has been in the index since 21 Sep 2026. How this is checked

RoboHarm is a Python package for BenchFlow, a robotic agent safety benchmark using paired harmful and benign evaluations. It is intended for researchers assessing manipulation-capable robotic agents, though the implementation is currently private and reserved for a future release.

Inferred · not functionally tested

Open SourceApache-2.0CLISelf-hosted
Visit PyPI

Overview

5 features

Purpose: Evaluating whether robotic agents safely distinguish and handle harmful versus benign tasks.

Inferred · not functionally tested

Audience: robotics and AI safety researchers

Inferred · not functionally tested

Functions: Unknown

Interfaces: API: unknown · MCP: unknown · CLI: indicated (inferred, not tested) · Self-hosting: indicated (inferred, not tested)

Recorded constraints: pricing: open_source · license: Apache-2.0 · platforms: CLI · deployment: cli, self_hosted

Constraint provenance is unknown; confirm requirements with the publisher.

Record sources: pypi.org. These links do not verify the individual claims.

In the Agent evaluation & testing space, roboharm takes a focused approach. Inferred · not functionally tested: It focuses on evaluating whether robotic agents safely distinguish and handle harmful versus benign tasks. Inferred · not functionally tested: It is built as an open-source project for robotics and AI safety researchers. Basis unknown · not verified: roboharm is open source under the Apache-2.0 license. Basis unknown · not verified: It ships for the command line, and it can be self-hosted.

roboharm first shipped in 2026. Inferred · not functionally tested: Key capabilities include paired evaluations, harmful task testing, and benign task testing. Access is currently waitlist-only.

Summary written by a language model from the project’s public pages.

Tasks: Inferred · not functionally tested

  • Paired evaluations
  • Harmful task testing
  • Benign task testing
  • Robotic agent benchmarking
  • Safety assessment

Topics: Inferred · not functionally tested

Tags
robotic-agentssafety-benchmarkharmful-benign-evaluationagent-evaluation
AI capabilities
Structured

JSON profile · Text profile · Access guide

Built with & integrations

Runs on
CLISelf-hosted

Trust & compliance

License
Apache-2.0
Public signals
HTTPSOpen SourceFree tier

Indexing history

3

What PulseGate has recorded for this listing

  1. Indexed27 Sep · 18:23 UTC
    roboharm seen via PyPI Fresh Feed
    Source: PyPI Fresh Feed · Open
  2. Indexed27 Sep · 17:51 UTC
    roboharm seen via PyPI Fresh Feed
    Source: PyPI Fresh Feed · Open
  3. Indexed21 Sep · 09:35 UTC
    roboharm seen via PyPI Bulk Enumerator
    Source: PyPI Bulk Enumerator · Open

Frequently asked questions about roboharm

What is roboharm?
Inferred · not functionally tested: Roboharm focuses on evaluating whether robotic agents safely distinguish and handle harmful versus benign tasks. It is catalogued under Agent evaluation & testing on PulseGate.
Who should use roboharm?
Inferred · not functionally tested: roboharm is an open-source project built for robotics and AI safety researchers.
Is roboharm free?
Basis unknown · not verified: Yes — roboharm is open source under the Apache-2.0 license and free to use.
What platforms does roboharm run on?
Basis unknown · not verified: roboharm runs on the command line. It can also be self-hosted.
Is roboharm still active?
PulseGate's liveness check found it on 21 Sep 2026.
How long has roboharm been around?
roboharm first shipped in 2026.
Is roboharm open source?
Basis unknown · not verified: Yes — roboharm is open source under the Apache-2.0 license.

Similar projects

Closest matches by what these projects do