Skip to content
Back to the index

Harnyx

PyPIInfrastructure

PulseGate's liveness check found it on 28 Sep 2026; it is registered on GitHub and PyPI and has been in the index since 28 Sep 2026. How this is checked

Harnyx is an Apache-2.0 Python package that implements Harness-R1, enabling executable AI-agent harnesses to improve from failure trajectories. It is intended for developers and researchers building and evaluating learning agent systems.

Inferred · not functionally tested

Open SourceApache-2.0CLISelf-hostedAPI
Visit PyPI

Overview

5 features

Purpose: Improving AI-agent harnesses by learning from failures and execution trajectories.

Inferred · not functionally tested

Audience: AI agent developers and machine learning researchers

Inferred · not functionally tested

Functions: agents

Inferred · not functionally tested

Interfaces: API: indicated (inferred, not tested) · 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, api_only

Constraint provenance is unknown; confirm requirements with the publisher.

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

Harnyx sits in PulseGate's Frameworks & runtimes category. Inferred · not functionally tested: It focuses on improving AI-agent harnesses by learning from failures and execution trajectories. Inferred · not functionally tested: It is built as an open-source project for AI agent developers and machine learning researchers. Basis unknown · not verified: The project is open source (Apache-2.0). Basis unknown · not verified: Harnyx is available on the command line and API, and it can be self-hosted.

Behind Harnyx is Nilay Mallik, and it first shipped in 2026. The project is developed in the open on GitHub with 15 commits in the last 90 days. Inferred · not functionally tested: Among its 5 catalogued features are failure trajectories, agent harness learning, and reinforcement learning.

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

Tasks: Inferred · not functionally tested

  • Failure trajectories
  • Agent harness learning
  • Reinforcement learning
  • Executable agents
  • Python package

Topics: Inferred · not functionally tested

Tags
agent-harnessesreinforcement-learningfailure-trajectoriesexecutable-agents
AI capabilities
TextCode
Inference: LocalWeights: Open

JSON profile · Text profile · Access guide

Built with & integrations

Runs on
CLISelf-hostedAPI-only

Trust & compliance

License
Apache-2.0
Public signals
HTTPSOpen SourceFree tierGitHubActive maintenance

Indexing history

2

What PulseGate has recorded for this listing

  1. Indexed28 Sep · 05:55 UTC
    harnyx seen via PyPI Fresh Feed
    Source: PyPI Fresh Feed · Open
  2. Indexed28 Sep · 05:39 UTC
    harnyx seen via PyPI Bulk Enumerator
    Source: PyPI Bulk Enumerator · Open

Frequently asked questions about Harnyx

What is Harnyx?
Inferred · not functionally tested: Harnyx focuses on improving AI-agent harnesses by learning from failures and execution trajectories. It is catalogued under Frameworks & runtimes on PulseGate.
Who is Harnyx for?
Inferred · not functionally tested: Harnyx is an open-source project built for AI agent developers and machine learning researchers.
Is Harnyx free?
Basis unknown · not verified: Yes — Harnyx is open source under the Apache-2.0 license and free to use.
What platforms does Harnyx run on?
Basis unknown · not verified: Harnyx runs on the command line and API. It can also be self-hosted.
Is Harnyx still maintained?
PulseGate's liveness check found it on 28 Sep 2026. Its GitHub repository shows 15 commits in the last 90 days.
What projects are similar to Harnyx?
Similar projects tracked by PulseGate include harnix, Harneloop, and harnify.harnixHarneloopharnify
Who develops Harnyx?
Harnyx is developed by Nilay Mallik.
How long has Harnyx been around?
Harnyx first shipped in 2026.

Similar projects

Closest matches by what these projects do