Who should use belay-harness?

belay-harness is an open-source project built for AI safety researchers and agent developers.

Is belay-harness free?

Yes — belay-harness is open source under the Apache-2.0 license and free to use.

What platforms does belay-harness run on?

belay-harness runs on the command line. It can also be self-hosted.

Is belay-harness still active?

The GitHub repository shows 87 commits in the last 90 days.

What are alternatives to belay-harness?

Similar projects tracked by PulseGate include constraintloop, trajectory-judge, and DeFake.

Who makes belay-harness?

belay-harness is developed by haqaliz.

How long has belay-harness been around?

belay-harness first shipped in 2026.

Back to the index

belay-harness

Name: belay-harness
Author: haqaliz

PyPIInfrastructure

belay-harness is an open-source CLI harness for AI agents, enabling sandboxed execution, step-by-step verification via replay, and deterministic trace logging. It is designed for developers and researchers focused on AI safety and agent reliability.

Open SourceApache-2.0CLISelf-hosted

Visit PyPI

Overview

5 features

In the LLM eval & observability space, belay-harness takes a focused approach. It focuses on ensuring safe and verifiable execution of AI agents by sandboxing and deterministic replay. It is built as an open-source project for AI safety researchers and agent developers. belay-harness is open source under the Apache-2.0 license. It runs on the command line, and it can be self-hosted.

haqaliz builds and maintains belay-harness, and it first shipped in 2026. Development happens publicly on GitHub with 87 commits in the last 90 days. Among its 5 catalogued features are agent sandboxing, step verification, and replay against real state.

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

Agent sandboxing
Step verification
Replay against real state
Deterministic trace
AI safety tools

Built with & integrations

Runs on

CLISelf-hosted

Trust & compliance

License

Apache-2.0

Verified signals

HTTPSOpen SourceFree tierGitHubActive maintenance

Indexing history

What PulseGate has recorded for this listing

Indexed18 Jul · 18:21 UTC
belay-harness verified against its public source
Source: PulseGate · Open

Frequently asked questions about belay-harness

What does belay-harness do?: Belay-harness focuses on ensuring safe and verifiable execution of AI agents by sandboxing and deterministic replay. It is catalogued under LLM eval & observability on PulseGate.
Who should use belay-harness?: belay-harness is an open-source project built for AI safety researchers and agent developers.
Is belay-harness free?: Yes — belay-harness is open source under the Apache-2.0 license and free to use.
What platforms does belay-harness run on?: belay-harness runs on the command line. It can also be self-hosted.
Is belay-harness still active?: The GitHub repository shows 87 commits in the last 90 days.
What are alternatives to belay-harness?: Similar projects tracked by PulseGate include constraintloop, trajectory-judge, and DeFake.constraintloop trajectory-judge DeFake
Who makes belay-harness?: belay-harness is developed by haqaliz.
How long has belay-harness been around?: belay-harness first shipped in 2026.

At a glance

Platforms: Cli
Languages: English
Open source: Yes (GitHub)
License: Apache-2.0
First seen: 18 Jul 2026
Built for: AI safety researchers and agent developers
Model: Open source
Solves: Ensuring safe and verifiable execution of AI agents by sandboxing and deterministic replay.

Registered as

GitHub: haqaliz/belay
PyPI: belay-harness

Developer

haqaliz

Solo developer

GitHub

Open source

View on GitHub

Stars: 0
Forks: 0
Open issues: 0
Last commit: 18 Jul 2026
Commits 90d: 87
Contributors: 1
Authorship: Solo
Default branch: master
Latest release: v0.1.0 · 18 Jul 2026

Live coverage

Identity confidence: Low · 64
Indexed: 18 Jul 2026
Lifecycle: Unverified
First seen: Jul 2026
Last seen: 18 Jul 2026

Identity audit (12)

Slug: belay-harness-pypi-org
Verification state: Indexed for public listing
Listing state: Listed: yes
Index status: Included in index
Latest evidence snapshot: 18 Jul 2026
Timeline basis: Indexed-at chronology (no inferred launch/funding milestones).
Name from: Written by a language model from the project's public pages.
Category from: Assigned by a language model.
Summary from: Written by a language model from public pages.
Languages from: Detected by a language model from page content.
Last updated: 25 Jul 2026
Canonical URL: https://pypi.org/project/belay-harness

Ship this? Send a correction — no account, and you get a link to follow it.

Also in LLM eval & observability

Same category — not a similarity match