Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

Get Started Free →

Field Watch Skill Builder · fengxia-pro | DecimalAI

agent toolingby fengxia-pro

Field Watch Skill Builder

fengxia-pro/field-watch-skill-builder

Source fengxia-pro/field-watch-skill-builder ↗License MIT

根据用户给出的任意研究领域、研究方向或交叉学术主题，生成可直接保存和安装的专属文献推送技能 SKILL.md。

SkillScore64Maintained

Automatic

Research pack

Last verified 8/4/2026

What it changes

Passed 14 of 22 test cases with the skill, and 6 of 22 without it.

Without skill27%

With skill64%

Checks it passed that the model alone missed

The generated SKILL.md contains a section detailing success criteria and prohibited items.
The generated SKILL.md starts with standard frontmatter including at least name and description keys.
The generated SKILL.md prohibits treating keyword presence alone as sufficient justification for pushing a paper.

and 9 more in the full report

9 fixed5 passed either way7 still failing1 broke

Pass-rate lift: +36 pp
Tokens: +121%
Agent turns: 0%
Response time: +41%

Model gemini-3.6-flashMeasured 8/4/2026v114 of 22 cases passed with skill

Tokens, turns and time describe resource use, not a dollar price.

Use this skill

Save it into your project's skills folder. Claude Code picks it up in its next session.

curl -fsSL --create-dirs -o .claude/skills/fengxia-pro-field-watch-skill-builder/SKILL.md https://app.decimal.ai/s/fengxia-pro-field-watch-skill-builder/SKILL.md

Want it pinned and updatable? pip install "decimalai>=0.15", then decimalai skills add fengxia-pro-field-watch-skill-builder, and start Claude Code with claude --plugin-dir "$(git rev-parse --show-toplevel)/.agents"

1 · One-time — install the CLI (docs ↗)

pip install "decimalai>=0.15"

2 · Get the skill — writes .agents/skills/fengxia-pro-field-watch-skill-builder/ and pins it in .decimal/skills.lock

decimalai skills add fengxia-pro-field-watch-skill-builder

No CLI? Get the same file — curl · raw file

curl -fsSL --create-dirs -o .claude/skills/fengxia-pro-field-watch-skill-builder/SKILL.md https://app.decimal.ai/s/fengxia-pro-field-watch-skill-builder/SKILL.md

Raw file: /s/fengxia-pro-field-watch-skill-builder/SKILL.md ↗ · machine-readable summary at /s/fengxia-pro-field-watch-skill-builder.json ↗

Safety checks: 3 of 3 passed

Code scan CleanIntent PassedContent Clean

See it in action

One of 9 test cases the skill fixed. Same model, same prompt, with and without the skill.

1 of 9

Prompt

请帮我针对“全固态钠离子电池电解质”这个研究方向，定制一份能够直接保存使用的文献推送技能文件。我需要你先给出领域界定的生成说明，推荐一个适合作为目录的英文短小写文件名，然后输出包含核心团队、匹配期刊、检索关键词、筛选标准和论文卡片模板的完整 SKILL.md 内容。

What was checked, without and with the skill
Checked	Without skill	With skill
The generated SKILL.md specifies daily search execution that checks the previous day and expands to 7 days if empty.	Missed	Passed
The generated SKILL.md includes the structured literature card template featuring core problem, solution, and logical premise sections.	Missed	Passed
The critical reading guidelines in SKILL.md emphasize battery science requirements like control experiments, characterization loops, and stability.	Missed	Passed

case-01 · gemini-3.6-flash · 8/4/2026 · answers are capped and redacted

What's inside

The instructions your model reads when the skill loads.

SKILL.md191 linessingle file

慢老师的领域文献推送技能生成器

角色定位

你是“慢老师的领域文献推送技能生成器”。你的任务不是直接推送文献，而是根据用户给出的任意研究领域、研究方向或学术主题，生成一个该领域专属、可直接保存为 SKILL.md 的文献推送技能。

生成结果必须是完整技能文件，而不是建议、摘要或半成品。默认使用简体中文，保留“慢老师文献推送卡片”的可读性、教学感和批判性阅读框架。

兼容性原则

生成的技能优先兼容 Codex / Claude 类 SKILL.md 技能格式：frontmatter 至少包含 name 和 description。
若需要 OpenClaw 兼容信息，把 version、user-invocable、metadata.openclaw 等放入 metadata 或在说明中提醒用户按平台要求迁移，避免破坏 Codex 基础校验。
不要依赖某个平台私有工具才能理解技能；检索工具、联网工具、数据库工具只能写成“可用时优先使用”。

输入理解

用户输入可能是单一领域、领域加子方向、交叉方向或含糊主题。优先做合理收敛，不要反复追问。

单一领域：直接生成该领域版本，例如“钠离子电池”“AI agent evaluation”。
领域加子方向：聚焦到子方向，例如“CO2RR 中的铜基催化剂”不能写成宽泛电催化。
交叉方向：同时覆盖两个学科的团队、期刊、关键词和争议，例如“AI for catalysis”。
含糊输入：先用 1-3 句说明工作假设，再生成可用版本，并在技能中写明适用范围。

领域分析流程

每次生成前，先完成以下判断，并把结论反映到最终技能中：

识别领域类型：实验科学、工程与应用科学、生命科学与医学、AI / 计算 / 数据科学、社会科学与教育，或交叉学科。
界定领域边界：说明研究对象、主线问题、常见子问题，以及不应混入的相邻内容。
筛选核心团队：给出 5-8 个近 5-10 年持续活跃、值得长期 follow 的团队或作者簇，尽量覆盖机制派、方法派、应用派、框架/综述派。新兴领域可放宽年限，但要说明。
构建期刊池：给出 6-10 本真实匹配该领域发表生态的期刊，分为核心期刊和扩展顶刊/邻近强刊。不要机械套用 Nature / Science / JACS / Angew。
生成关键词组：给出 3-5 组英文检索词，必要时补中文说明。关键词要适合高信噪比监控，避免过宽、过虚。
提炼关键争议：总结 3-5 个当前最值得盯的机制、评价、benchmark、可重复性、外部有效性、工程放大或转化争议。
制定纳入与排除标准：明确哪些论文应纳入，哪些只是词面相关但属于噪音。

证据与诚实性

有联网、文献数据库、搜索或浏览工具时，优先核验团队、期刊和关键词生态；尤其要核验近 5 年活跃度和是否仍在该方向持续发表。
没有检索工具时，可以基于已有知识生成，但必须把不确定边界写进“生成说明”或技能正文中的“定期复核”要求。
不得编造不存在的团队、期刊、数据库、benchmark 或学科结构。

Track record

Benchmark

14 / 22

test cases passed, vs 6 / 22 without

Adoption

Users: 0
Times used: 0

Safety and source details

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 8/1/2026

View 15 scanner checks

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

fengxia-pro / field-watch-skill-builder

field-watch-skill-builder/SKILL.md

View on GitHub ↗

License MITImported 8/1/2026Version v1

More skills from the same source and nearby in the registry, with their headline benchmark lift.