leaderboard
Track AI model leaderboard rankings over time and detect rank/score changes between snapshots
- License 许可
- Apache-2.0
- 最近更新
- 2026-11-14
- Stars
- ★ 1.6k
- Forks
- 491
01 前言 Front Matter
- name
- leaderboard
- version
- 1.3.0
- description
- Track AI model leaderboard rankings over time and detect rank/score changes between snapshots
02 资源文件 1
- 脚本scripts/leaderboard_snapshot.py1.7 KB
03 SKILL.md 正文
Leaderboard Snapshot Tracker
Track AI model leaderboard rankings over time using SQLite snapshots. Detect new models, removed models, rank changes, and score changes between dates.
Supported Leaderboards
| Leaderboard | URL | Modality |
|---|---|---|
| LMSYS Chatbot Arena | https://lmsys.org | Text, Vision |
| LMArena | https://lmarena.ai | Text, Vision |
| HuggingFace Open LLM | https://huggingface.co/spaces/open-llm-leaderboard | Text |
| Artificial Analysis | https://artificialanalysis.ai | Text, Image, Video |
| Scale AI SEAL | https://scale.com/leaderboard | Text |
Usage
Save a snapshot
cd {SKILL_DIR} && python3 skills/leaderboard/scripts/leaderboard_snapshot.py save \
--leaderboard "chatbot-arena" \
--date 2026-04-14 \
--data '[{"model": "claude-4-opus", "rank": 1, "score": 1350}]'
Prints the diff against the previous snapshot (new models, rank changes, score changes).
View latest snapshot
cd {SKILL_DIR} && python3 skills/leaderboard/scripts/leaderboard_snapshot.py latest \
--leaderboard "chatbot-arena"
Data Storage
Snapshots are stored in ~/.cache/morning-ai/leaderboard.db (SQLite). Each entry has:
leaderboard— leaderboard identifiermodel— model namerank— position on the leaderboardscore— numeric score (ELO, accuracy, etc.)snapshot_date— date of the snapshot
Integration with Main Workflow
This skill is currently a standalone utility. It can be integrated into the main morning-ai workflow as a Benchmark data source:
- As a collector: Scrape leaderboard pages → save snapshot → diff against previous → generate
TrackerItementries for rank changes - As a report section: Add a "Leaderboard Movement" section to the daily report showing rank/score deltas
To integrate, a collector module (lib/leaderboard_collector.py) would:
- Fetch current leaderboard data from supported sites
- Call
save_snapshot()to persist - Call
diff_snapshot()to detect changes - Convert significant changes (new #1 model, big rank jumps) into
TrackerItemobjects with Benchmark type
04 同仓库更多 Skills
worktable-review-with-annotations
Create, inspect, reply to, update, or resolve situated Worktable annotations. Use when feedback belongs to a Doc section, block, text range, or HTML Doc, or when the user asks to address existing review feedback and coordinate a requested artifact edit.
worktable-find-and-synthesize-context
Find, assess, and synthesize context from Worktable Docs and Records. Use when the user asks what a workspace contains, wants workspace-backed research, needs current facts reconciled across artifacts, or requests a cited synthesis without changing workspace state.
worktable-create-or-update-docs
Create, edit, rename, or maintain narrative Worktable Docs. Use for plans, decisions, briefs, notes, research syntheses, and other durable prose when the user requests a new artifact or a safe update to an existing one.
worktable-create-or-manage-records
Organize repeated structured information as durable Worktable Records. Use for research datasets, inventories, directories, pipelines, trackers, and other collections whose items need identity, fields, filtering, links, validation, or individual updates.
worktable-create-or-manage-interactive-html
Create or improve durable visual and interactive artifacts in Worktable. Use for dashboards, trackers, calculators, mini apps, slide presentations, and other requests that benefit from a polished HTML Doc.
worktable-collaborate-threads
Read, answer, and hand off messages through durable Worktable threads with users or other authenticated participants. Use when the user asks to check addressed messages, reply to a participant, continue a thread, request another agent's help, wait briefly for a response, or watch for comments and messages sent to the agent.