From 7a1571f46b72dd1ed7592e2da9fb60c958c507fb Mon Sep 17 00:00:00 2001 From: Jesse Zhang <101612750+Nagi-ovo@users.noreply.github.com> Date: Sun, 14 Jun 2026 00:16:49 +0100 Subject: [PATCH] Remove empty artifacts from Cursor runs --- README.en.md | 5 ++-- README.md | 5 ++-- models/claude-opus-4-6-cursor.json | 7 ------ models/claude-sonnet-4-6-cursor.json | 24 ------------------- models/gemini-3-1-pro-cursor.json | 7 ------ .../cursor-max/pelican-cycling.svg | 0 .../cursor-high/mythos-craft.html | 0 .../cursor-high/pelican-cycling.svg | 0 .../cursor-default/pelican-cycling.svg | 0 scripts/validate-data.ts | 7 ++++-- 10 files changed, 9 insertions(+), 46 deletions(-) delete mode 100644 models/claude-sonnet-4-6-cursor.json delete mode 100644 outputs/claude-opus-4-6/cursor-max/pelican-cycling.svg delete mode 100644 outputs/claude-sonnet-4-6/cursor-high/mythos-craft.html delete mode 100644 outputs/claude-sonnet-4-6/cursor-high/pelican-cycling.svg delete mode 100644 outputs/gemini-3-1-pro/cursor-default/pelican-cycling.svg diff --git a/README.en.md b/README.en.md index 50f7700..13758d2 100644 --- a/README.en.md +++ b/README.en.md @@ -29,11 +29,10 @@ One-shot LLM eval cases by NAGI STUDIO: same prompt, different model x harness x | Claude Fable 5 | Anthropic | Claude Web App · Max
Claude Code · Max
Cursor · High | 06 | | Claude Opus 4.8 | Anthropic | Claude Code · Max | 02 | | Claude Opus 4.7 | Anthropic | Cursor · Max | 02 | -| Claude Opus 4.6 | Anthropic | Cursor · Max | 02 | -| Claude Sonnet 4.6 | Anthropic | Cursor · High | 02 | +| Claude Opus 4.6 | Anthropic | Cursor · Max | 01 | | GPT-5.5 Pro | OpenAI | ChatGPT Web · Extended Pro | 02 | | GPT-5.5 | OpenAI | Codex CLI · xhigh | 02 | -| Gemini 3.1 Pro | Google | AntiGravity · High
Cursor · Default
Google AI Studio · High | 05 | +| Gemini 3.1 Pro | Google | AntiGravity · High
Cursor · Default
Google AI Studio · High | 04 | | Gemini 3.5 Flash | Google | AntiGravity · High
Google AI Studio · High
Cursor · Default | 04 | | Grok Build | xAI | Grok Build TUI · Max | 02 | | Composer 2.5 | Cursor | Cursor · Max
Cursor · Default
Grok Build TUI · Default | 05 | diff --git a/README.md b/README.md index 5db3170..d8635c7 100644 --- a/README.md +++ b/README.md @@ -29,11 +29,10 @@ NAGI STUDIO 的 LLM 测评案例集:同一段提示词,不同「模型 × Ha | Claude Fable 5 | Anthropic | Claude Web App · Max
Claude Code · Max
Cursor · High | 06 | | Claude Opus 4.8 | Anthropic | Claude Code · Max | 02 | | Claude Opus 4.7 | Anthropic | Cursor · Max | 02 | -| Claude Opus 4.6 | Anthropic | Cursor · Max | 02 | -| Claude Sonnet 4.6 | Anthropic | Cursor · High | 02 | +| Claude Opus 4.6 | Anthropic | Cursor · Max | 01 | | GPT-5.5 Pro | OpenAI | ChatGPT Web · Extended Pro | 02 | | GPT-5.5 | OpenAI | Codex CLI · xhigh | 02 | -| Gemini 3.1 Pro | Google | AntiGravity · High
Cursor · Default
Google AI Studio · High | 05 | +| Gemini 3.1 Pro | Google | AntiGravity · High
Cursor · Default
Google AI Studio · High | 04 | | Gemini 3.5 Flash | Google | AntiGravity · High
Google AI Studio · High
Cursor · Default | 04 | | Grok Build | xAI | Grok Build TUI · Max | 02 | | Composer 2.5 | Cursor | Cursor · Max
Cursor · Default
Grok Build TUI · Default | 05 | diff --git a/models/claude-opus-4-6-cursor.json b/models/claude-opus-4-6-cursor.json index a37ac31..d42bccb 100644 --- a/models/claude-opus-4-6-cursor.json +++ b/models/claude-opus-4-6-cursor.json @@ -12,13 +12,6 @@ "en": "Generated in one shot by Claude Opus 4.6 (Max effort) inside Cursor; unmodified original output." }, "contributor": "Nagi-ovo" - }, - "pelican-cycling": { - "note": { - "zh": "在 Cursor 中以 Claude Opus 4.6(Max 思考配额)一次生成;本页为未经修改的原始产出。", - "en": "Generated in one shot by Claude Opus 4.6 (Max effort) inside Cursor; unmodified original output." - }, - "contributor": "Nagi-ovo" } } } diff --git a/models/claude-sonnet-4-6-cursor.json b/models/claude-sonnet-4-6-cursor.json deleted file mode 100644 index 7ad5766..0000000 --- a/models/claude-sonnet-4-6-cursor.json +++ /dev/null @@ -1,24 +0,0 @@ -{ - "label": "Claude Sonnet 4.6", - "vendor": "Anthropic", - "harness": "Cursor", - "effort": "High", - "artifactDir": "claude-sonnet-4-6/cursor-high", - "order": 18, - "runs": { - "mythos-craft": { - "note": { - "zh": "在 Cursor 中以 Claude Sonnet 4.6(High 思考配额)一次生成;本页为未经修改的原始产出。", - "en": "Generated in one shot by Claude Sonnet 4.6 (High effort) inside Cursor; unmodified original output." - }, - "contributor": "Nagi-ovo" - }, - "pelican-cycling": { - "note": { - "zh": "在 Cursor 中以 Claude Sonnet 4.6(High 思考配额)一次生成;本页为未经修改的原始产出。", - "en": "Generated in one shot by Claude Sonnet 4.6 (High effort) inside Cursor; unmodified original output." - }, - "contributor": "Nagi-ovo" - } - } -} diff --git a/models/gemini-3-1-pro-cursor.json b/models/gemini-3-1-pro-cursor.json index 98b796c..b793f10 100644 --- a/models/gemini-3-1-pro-cursor.json +++ b/models/gemini-3-1-pro-cursor.json @@ -12,13 +12,6 @@ "en": "Generated in one shot by Gemini 3.1 Pro inside Cursor with default settings; unmodified original output." }, "contributor": "Nagi-ovo" - }, - "pelican-cycling": { - "note": { - "zh": "在 Cursor 中以 Gemini 3.1 Pro 默认设置一次生成;本页为未经修改的原始产出。", - "en": "Generated in one shot by Gemini 3.1 Pro inside Cursor with default settings; unmodified original output." - }, - "contributor": "Nagi-ovo" } } } diff --git a/outputs/claude-opus-4-6/cursor-max/pelican-cycling.svg b/outputs/claude-opus-4-6/cursor-max/pelican-cycling.svg deleted file mode 100644 index e69de29..0000000 diff --git a/outputs/claude-sonnet-4-6/cursor-high/mythos-craft.html b/outputs/claude-sonnet-4-6/cursor-high/mythos-craft.html deleted file mode 100644 index e69de29..0000000 diff --git a/outputs/claude-sonnet-4-6/cursor-high/pelican-cycling.svg b/outputs/claude-sonnet-4-6/cursor-high/pelican-cycling.svg deleted file mode 100644 index e69de29..0000000 diff --git a/outputs/gemini-3-1-pro/cursor-default/pelican-cycling.svg b/outputs/gemini-3-1-pro/cursor-default/pelican-cycling.svg deleted file mode 100644 index e69de29..0000000 diff --git a/scripts/validate-data.ts b/scripts/validate-data.ts index dc53427..8fb1de3 100644 --- a/scripts/validate-data.ts +++ b/scripts/validate-data.ts @@ -1,7 +1,7 @@ // Validates the contribution data: models/.json registrations and // their artifacts under outputs/. Runs in CI for every PR and push, so a // malformed or undocumented contribution fails before it can ship. -import { existsSync, readFileSync, readdirSync } from 'node:fs' +import { existsSync, readFileSync, readdirSync, statSync } from 'node:fs' import { dirname, join } from 'node:path' import { fileURLToPath } from 'node:url' @@ -99,8 +99,11 @@ for (const f of readdirSync(modelsDir).filter((f) => f.endsWith('.json'))) { seenFiles.add(file) const artifact = `${artifactDir}/${file}` declaredArtifacts.add(artifact) - if (!existsSync(join(outputsDir, artifactDir, file))) { + const artifactPath = join(outputsDir, artifactDir, file) + if (!existsSync(artifactPath)) { errors.push(`models/${f}: ${where} declared but artifact missing: outputs/${artifact}`) + } else if (statSync(artifactPath).size === 0) { + errors.push(`models/${f}: ${where} artifact is empty: outputs/${artifact}`) } }) }