Skip to content

anbeime/skill / skills/cy-carousel-skill/cy-carousel-skill

cy-carousel

Verified

制作 AI 科技类小红书图文轮播(封面+10张内页,1440×1920,暗底或浅色纸底,电子青强调色):讲一个 AI 概念、一篇论文或一个产品,体裁是「AI 学习笔记」。查证一手出处、找开放授权素材(论文原文、真实截图、人物照),用 scripts/手排.py 照 ELIZA 的做法逐页手排(毛笔双色标题、原文纸条放大加红圈、电子青大数字、问句贴条),再用 render.py 渲染、check_all.py 检查、package.py 打包。用户要做 AI/科技/论文/产品类图文,或复用、精修这类轮播时使用。

FileKindSize
.github/ISSUE_TEMPLATE/bug_report.ymlconfig1.4 KB
.github/ISSUE_TEMPLATE/config.ymlconfig27 B
.github/ISSUE_TEMPLATE/feature_request.ymlconfig932 B
.github/agent/implement.mdmarkdown1.9 KB
.github/agent/review.mdmarkdown1.5 KB
.github/agent/triage.mdmarkdown2.9 KB
.github/labels.jsonconfig2.2 KB
.github/pull_request_template.mdmarkdown409 B
.github/scripts/apply_triage.pyscript6.4 KB
.github/scripts/dump_issue.shscript509 B
.github/scripts/intake_feedback.pyscript4.0 KB
.github/scripts/open_agent_pr.shscript3.8 KB
.github/workflows/agent-cleanup.ymlconfig1.9 KB
.github/workflows/agent-implement.ymlconfig4.5 KB
.github/workflows/agent-review.ymlconfig2.8 KB
.github/workflows/agent-triage.ymlconfig4.6 KB
.github/workflows/claude.ymlconfig2.9 KB
.github/workflows/feedback-intake.ymlconfig2.0 KB
.github/workflows/sync-labels.ymlconfig798 B
.github/workflows/validate.ymlconfig1.1 KB
.gitignoreother65 B
CLAUDE.mdmarkdown2.3 KB
LICENSEother1.2 KB
README.mdmarkdown10.6 KB
SKILL.mdskill md9.7 KB
assets/banner.jpgbinary422.3 KB
assets/fonts/LXGWWenKai-Medium.ttfbinary24.2 MB
assets/fonts/MaShanZheng-Regular.ttfbinary5.6 MB
assets/fonts/NotoSansSC[wght].ttfbinary16.9 MB
assets/fonts/NotoSerifSC[wght].ttfbinary24.0 MB
assets/fonts/README.mdmarkdown677 B
assets/fonts/licensed/README.mdmarkdown1.0 KB
assets/fonts/lxgwwenkai-OFL.txtmarkdown5.0 KB
assets/fonts/mashanzheng-OFL.txtmarkdown4.3 KB
assets/fonts/notosanssc-OFL.txtmarkdown4.3 KB
assets/fonts/notoserifsc-OFL.txtmarkdown4.2 KB
assets/fonts/来源与许可证.jsonconfig1.3 KB
docs/agent-workflow.mdmarkdown7.8 KB
examples/01-ELIZA/README.mdmarkdown846 B
examples/01-ELIZA/全套预览.jpgbinary408.3 KB
examples/01-ELIZA/标题.txtmarkdown67 B
examples/01-ELIZA/正文.txtmarkdown1.7 KB
examples/02-Dots/README.mdmarkdown875 B
examples/02-Dots/全套预览.jpgbinary408.9 KB
examples/02-Dots/标题.txtmarkdown69 B
examples/02-Dots/正文.txtmarkdown1.8 KB
examples/03-聪明汉斯/README.mdmarkdown1.1 KB
examples/03-聪明汉斯/全套预览.jpgbinary450.1 KB
examples/03-聪明汉斯/制作.pyscript14.2 KB
examples/04-AI幻觉/README.mdmarkdown1.4 KB
examples/04-AI幻觉/全套预览.jpgbinary668.3 KB
examples/04-AI幻觉/制作.pyscript12.7 KB
examples/04-AI幻觉/标题.txtmarkdown131 B
examples/04-AI幻觉/正文.txtmarkdown1010 B
references/排字细节.mdmarkdown6.9 KB
requirements.txtmanifest91 B
scripts/blind_test.pyscript3.6 KB
scripts/build.pyscript43.6 KB
scripts/check_all.pyscript13.3 KB
scripts/cutout.swiftscript858 B
scripts/emoji.swiftscript1.6 KB
scripts/io_paths.pyscript1.0 KB
scripts/layouts_v2.pyscript67.0 KB
scripts/ocr.swiftscript1.3 KB
scripts/ocrfind.swiftscript1.3 KB
scripts/package.pyscript4.0 KB
scripts/render.pyscript47.9 KB
scripts/保脸裁切.pyscript2.1 KB
scripts/图像等比.pyscript3.2 KB
scripts/对照.pyscript12.3 KB
scripts/手排.pyscript26.9 KB
scripts/检测人脸.swiftscript935 B
scripts/浅色排版.pyscript10.3 KB
scripts/验证排版.pyscript23.8 KB
scripts/验证留白.pyscript1.4 KB
scripts/验证背景.pyscript17.5 KB
tests/test_agent_workflow.pyscript7.4 KB
tests/test_cache_and_config.pyscript6.1 KB
tests/test_cross_platform.pyscript5.1 KB
tests/test_render_smoke.pyscript3.8 KB
tests/test_strict_visual.pyscript10.6 KB
tests/test_strict_visual_integration.pyscript6.7 KB
tools/submit_feedback.pyscript2.6 KB
tools/validate_repo.pyscript4.5 KB

.github/scripts/apply_triage.py

View on GitHub
#!/usr/bin/env python3
"""Turn the triage agent's JSON verdict into labels, a comment and an implement decision.

The agent only writes a JSON file; every write to GitHub happens here, so issue
text can never steer what gets labelled or whether code generation starts.
"""

import argparse
import json
import os
import re
import subprocess
import sys
from pathlib import Path


DECISIONS = {"accepted", "needs-info", "rejected", "duplicate"}
TYPES = {"bug", "feature", "docs", "question"}
SIZES = {"S", "M", "L"}
RISKS = {"low", "medium", "high"}
TRUSTED_ASSOCIATIONS = {"OWNER", "MEMBER", "COLLABORATOR"}
# 由维护者控制的入口(repository_dispatch / 手动登记)创建的反馈视为已过一道人工筛选。
TRUSTED_SOURCES = {"community", "social"}
MANAGED_PREFIXES = ("triage:", "type:", "size:", "risk:")
MANAGED_LABELS = {"needs-triage", "status:awaiting-maintainer"}
MARKER = "<!-- agent-triage -->"
SECRET_ENV = ("ANTHROPIC_API_KEY", "CLAUDE_CODE_OAUTH_TOKEN", "GH_TOKEN", "AGENT_GITHUB_TOKEN")
SECRET_PATTERN = re.compile(r"(sk-ant-[A-Za-z0-9_-]{8,}|gh[pousr]_[A-Za-z0-9]{20,}|github_pat_[A-Za-z0-9_]{20,})")


def redact(text):
    """Never echo credentials into a public comment, whatever the agent wrote."""
    for name in SECRET_ENV:
        value = os.environ.get(name)
        if value and len(value) >= 8:
            text = text.replace(value, "[REDACTED]")
    return SECRET_PATTERN.sub("[REDACTED]", text)


def load_result(path):
    data = json.loads(Path(path).read_text(encoding="utf-8"))
    decision = data.get("decision")
    if decision not in DECISIONS:
        raise ValueError(f"decision 无效: {decision!r}")
    data["type"] = data.get("type") if data.get("type") in TYPES else "question"
    data["size"] = data.get("size") if data.get("size") in SIZES else "L"
    data["risk"] = data.get("risk") if data.get("risk") in RISKS else "high"
    for key in ("plan", "questions"):
        value = data.get(key) or []
        data[key] = [str(item) for item in value] if isinstance(value, list) else [str(value)]
    duplicate = data.get("duplicate_of")
    data["duplicate_of"] = duplicate if isinstance(duplicate, int) and duplicate > 0 else None
    return data


def should_implement(result, association, source, auto_implement):
    return (
        auto_implement
        and result["decision"] == "accepted"
        and result["type"] in {"bug", "feature", "docs"}
        and result["size"] in {"S", "M"}
        and result["risk"] != "high"
        and (association in TRUSTED_ASSOCIATIONS or source in TRUSTED_SOURCES)
    )


def render_comment(result, implement):
    lines = [MARKER, "### 🤖 Agent 需求评估", ""]
    verdict = {
        "accepted": "✅ 可行,建议采纳",
        "needs-info": "❓ 信息不足,需要补充",
        "rejected": "⛔ 暂不采纳",
        "duplicate": "🔁 疑似重复",
    }[result["decision"]]
    lines.append(
        f"**结论**:{verdict} · 类型 `{result['type']}` · 规模 `{result['size']}` · 风险 `{result['risk']}`"
    )
    if result.get("summary"):
        lines += ["", f"**需求摘要**:{result['summary']}"]
    if result.get("reasoning"):
        lines += ["", f"**理由**:{result['reasoning']}"]
    if result["duplicate_of"]:
        lines += ["", f"**可能重复**:#{result['duplicate_of']}"]
    if result["plan"]:
        lines += ["", "**实现思路**:"] + [f"{i}. {step}" for i, step in enumerate(result["plan"], 1)]
    if result["questions"]:
        lines += ["", "**需要补充的信息**:"] + [f"- {q}" for q in result["questions"]]
    lines.append("")
    if implement:
        lines.append("➡️ 已自动进入编码阶段,Agent 完成后会提交 PR 并关联本 issue。")
    elif result["decision"] == "accepted":
        lines.append("⏸ 等待维护者确认:添加 `agent:implement` 标签即可让 Agent 开始编码。")
    elif result["decision"] == "needs-info":
        lines.append("补充信息后编辑 issue 或留言,维护者可重新触发评估。")
    else:
        lines.append("以上为自动评估,最终是否关闭由维护者决定。")
    return "\n".join(lines)


def plan_actions(result, association, source, auto_implement):
    implement = should_implement(result, association, source, auto_implement)
    labels = [
        f"triage:{result['decision']}",
        f"type:{result['type']}",
        f"size:{result['size']}",
        f"risk:{result['risk']}",
    ]
    if result["decision"] == "accepted" and not implement:
        labels.append("status:awaiting-maintainer")
    return {
        "labels": labels,
        "comment": redact(render_comment(result, implement)),
        "implement": implement,
    }


def gh(*args, check=True):
    return subprocess.run(["gh", *args], check=check, text=True, capture_output=True)


def apply(issue, actions):
    current = json.loads(gh("issue", "view", str(issue), "--json", "labels").stdout)["labels"]
    stale = [
        label["name"]
        for label in current
        if label["name"].startswith(MANAGED_PREFIXES) or label["name"] in MANAGED_LABELS
    ]
    stale = [name for name in stale if name not in actions["labels"]]
    if stale:
        gh("issue", "edit", str(issue), "--remove-label", ",".join(stale))
    gh("issue", "edit", str(issue), "--add-label", ",".join(actions["labels"]))
    gh("issue", "comment", str(issue), "--body", actions["comment"])


def main(argv=None):
    parser = argparse.ArgumentParser(description=__doc__)
    parser.add_argument("--result", required=True)
    parser.add_argument("--issue", required=True, type=int)
    parser.add_argument("--association", default="NONE")
    parser.add_argument("--source", default="github")
    parser.add_argument("--auto-implement", default="false")
    parser.add_argument("--dry-run", action="store_true")
    args = parser.parse_args(argv)

    result = load_result(args.result)
    actions = plan_actions(
        result,
        args.association.upper(),
        args.source,
        args.auto_implement.lower() == "true",
    )
    if args.dry_run:
        print(json.dumps(actions, ensure_ascii=False, indent=2))
    else:
        apply(args.issue, actions)

    output = os.environ.get("GITHUB_OUTPUT")
    if output:
        with open(output, "a", encoding="utf-8") as handle:
            handle.write(f"implement={'true' if actions['implement'] else 'false'}\n")
            handle.write(f"decision={result['decision']}\n")
    return 0


if __name__ == "__main__":
    sys.exit(main())
slack-gif-creatoranthropics/skillsKnowledge and utilities for creating animated GIFs optimized for Slack. Provides constraints, validation tools, and animation concepts. Use when users request animated GIFs for Slack like "make me a GIF of X doing Y for Slack."Stars: 197kUpdated Claude CodeCodexVerifiedInstall
ponytailDietrichGebert/ponytailLazy senior dev mode for any coding task: the smallest change that fully solves it, and a reply a busy human understands in one read.Stars: 194kUpdated Claude CodeCodexVerifiedInstall
ponytail-debtDietrichGebert/ponytailHarvest every shortcut: comment into one debt ledger, so deferrals get tracked instead of forgotten. One-shot report.Stars: 194kUpdated Claude CodeCodexVerifiedInstall
ponytail-debtDietrichGebert/ponytailList every `shortcut:` comment (and older `ponytail:` ones) as a debt ledger. One-shot report, changes nothing. Use for "ponytail debt", "what did ponytail defer", "list the shortcuts", /ponytail-debt.Stars: 194kUpdated Claude CodeCodexVerifiedInstall