run-hakari-benchmarks

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/hakari-bench-run-hakari-benchmarks

About

Use cases: frontend, ml, agent-building

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools

No tools found

Secrets

None asked for

Install scripts

None found

What's inside

Skill name

run-hakari-benchmarks

Description

Entry point for HAKARI-Bench evaluation work. Use when Codex is asked to benchmark or evaluate embedding models on NanoIR/Nano* datasets, choose prompt/truncate-dim/attention options, schedule GPU benchmark jobs, compare BM25, or refresh DuckDB and HTML leaderboard reports.

License

MIT

Instructions

19 lines in SKILL.md · view on GitHub

# Run HAKARI-Bench Evaluations

The canonical evaluation procedure lives in
[`docs/evaluation_policy.md`](../../docs/evaluation_policy.md).

Before running or scheduling evaluations, read that document and follow it as
the source of truth for:

- model research,
- prompt and attention choices,
- dense, sparse, late-interaction, reranker, and BM25 commands,
- embedding variant policy,
- coverage audits before reporting results,
- DuckDB/HTML viewer refresh expectations.

Do not maintain evaluation command details in this skill file. Update
`docs/evaluation_policy.md` instead.

Signals

Stars

34

Forks

1

Open issues

0

Last commit (this folder)

Jun 9, 2026 (4 months ago)

From GitHub (fetched Oct 24, 2026).

Details

License

MIT

Repository

github.com/hakari-bench/hakari-bench

Website

huggingface.co/spaces/hakari-bench/leaderboard

First seen

Oct 24, 2026

Last checked

Oct 24, 2026

Source

GitHubgithub-graphql