ai-benchmark-expert SKILL.md skillA
ai-benchmark-expert from andersonlimahw/lemon-ai-hub is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.
mdr add andersonlimahw/lemon-ai-hub/ai-benchmark-expert@git:20261004.8445980Pins git:20261004.8445980 in mdr.lock. Pin the fingerprint instead.
- Audit grade
- A, 17 of 17 checks passed
- Kind
- SKILL.md skill
- Repository
- andersonlimahw/lemon-ai-hub
- Path
plugins/ai-benchmark-expert/SKILL.md- Versions recorded
- 1
- Latest version
- git:20261004.8445980, committed Oct 4, 2026
- Size
- 6,991 bytes
- Repository stars
- 17
- License
- MIT
Benchmark LLMs by giving the SAME task to multiple providers and models as headless CLI subagents (Claude Code, Codex, Gemini CLI, OpenCode/MiniMax, mm-run or any command), scoring each with hidden deterministic checks plus a blind LLM judge on a weighted 8-dimension rubric, and publishing one self-contained interactive HTML report (leaderboard + tiers, quality-vs-cost Pareto, efficiency, rubric heatmap, sortable table, per-run evidence, methodology). Use when the user asks to benchmark…
Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.
What the file says
# AI Benchmark Expert Same task → many models → evidence → one shareable report. `scripts/bench.py` runs every **target** (a provider/model behind a headless CLI) on every **task** in its own isolated git workdir, executes hidden **checks** there, optionally asks a **blind judge** to score the anonymised output on a rubric, then writes `results.json` and a single-file `report.html` (no CDN, no network — attach it to a PR, mail it, host it). Methodology is modelled on public benchmarks — read [`references/methodology.md`](references/methodology.md) before designing a suite or interpreting close results. ## Workflow 1. **Frame the question.** What decision will the benchmark inform (pick a default coding model? a cheap subagent tier? a provider for long context)? That picks the tasks. A benchmark answers *this suite*, not "best model". 2. **Write the suite** (`suite.json`). Start from [`examples/quick-suite.json`](examples/quick-suite.json). Each task needs a precise prompt and at least one deterministic check; checks are what make the score honest. Put acceptance scripts **outside** the workdir (use `$BENCH_SUITE_DIR`) so models cannot read or edit them. …
Read the whole file at its exact version.
How to install
mdr add andersonlimahw/lemon-ai-hub/ai-benchmark-expert@sha256:88c0d2777eaf462fThe line at the top of this page pins the label git:20261004.8445980, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.
[](https://markdownregistry.com/a/art_nxbn2t2ynlnzrjty)
1 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 1 version of this file, committed on Oct 4, 2026.
Audit of the latest version
- pass: Frontmatter block present
- pass: Frontmatter declares a name
- pass: Frontmatter declares a description
- pass: Size between 200 bytes and 200 KB (6991 bytes)
- pass: No zero-width or bidi control characters
- pass: No instruction hidden inside an HTML comment
- pass: No link to an exfiltration or paste host
- pass: No credential-shaped string
- pass: No instruction to send local credentials anywhere
- pass: No text hidden with inline styles
- pass: No prompt-injection phrasing
- pass: No curl or wget piped into a shell
- pass: No recursive delete of root, home or parent
- pass: No instruction to read or print local credentials
- pass: No base64 blob over 200 characters
- pass: No link to a raw IP address
- pass: No script tag
Source
andersonlimahw/lemon-ai-hub · 17 stars · license MIT · pushed 2026-10-16 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_nxbn2t2ynlnzrjty GET https://markdownregistry.com/api/v1/resolve?ref=andersonlimahw/lemon-ai-hub/ai-benchmark-expert GET https://markdownregistry.com/api/v1/blob/88c0d2777eaf462fae794937eb2bff52befcb74cc1828035ca4fe31fafd0c33c
Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.
Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.