Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add instructions/light0305/light-skills/agents-mdgit clone --depth 1 https://github.com/Light0305/Light-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/light0305/light-skills/agents-md)<a href="https://agentmods.dev/instructions/light0305/light-skills/agents-md"><img src="https://agentmods.dev/badge/instructions/light0305/light-skills/agents-md.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00548 | $0.00548 |
| Opus 5 | $0.00274 | $0.00274 |
| Sonnet 5 | $0.00110 | $0.00110 |
| Haiku 4.5 | $0.00055 | $0.00055 |
Grade A, and why
Light-skills AGENTS.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 5d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
Light-Skills 代理协作说明
本仓库是公开科研全流程技能包，面向 Claude Code / Codex / OpenCode。请把 skills/ 视为唯一 canonical 技能源；不要把维护者私人记忆、对话史、邮箱、未公开研究、db01-db09、MCP 配置或本地知识库内容写进仓库。
客户端发现路径
三类客户端默认发现路径不同。项目级使用前先运行：
$env:PYTHONUTF8="1"
python scripts\bootstrap_agent_skills.py --mode auto --force
该脚本把 skills/ 安全映射到：
.agents/skills：Codex / OpenCode open-standard 项目技能路径。.claude/skills：Claude Code 项目技能路径，也可被 OpenCode 读取。.opencode/skills：OpenCode 原生项目技能路径。
Windows 不能创建目录 symlink 时，脚本会回退为 copy，并写入 .light_generated_link 标记；修改 canonical skills/ 后需重新运行 --force 刷新生成副本。脚本不会覆盖没有 Light 生成标记的用户目录。
开发纪律
- commit 只署作者本人；标题和正文写中文；不要添加 Co-Authored-By。
- 不要 push，除非用户明确要求。
- Python 命令前设置
PYTHONUTF8=1；临时产物放.upgrade/_e2e。 - 用户面向脚本必须亲手跑
--selftest到 exit 0，不能相信“已完成”报告。 - 动态事实、论文、DOI、star、链接、能力声明查不到就写
unknown/待核查。 - 论文数据图必须程序化生成；禁止 AI 生图冒充数据图。
- R/ggplot2 是增强路径，不是静默前提：缺 R 或缺包时，先给用户“继续 matplotlib 诚实降级 / 安装或配置 R / 提供 Rscript 路径”的选择；非交互自动流程默认降级并标
degraded=True，不得擅自安装。
快速自测
$env:PYTHONUTF8="1"
python scripts\bootstrap_agent_skills.py --selftest
python scripts\bootstrap_agent_skills.py --check-only
python skills\light-figure\scripts\r_ggplot.py --selftest
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 5d ago First seen · 40 lines · 548 tokens per session scan A 2312d0eff162
Light-skills AGENTS.md is an instructions file published in the GitHub repository Light0305/Light-skills (581 stars, last pushed 2mo ago), licensed MIT. It adds 548 tokens to every session, about $0.0027 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
paper-writing-skill CLAUDE.md
Instructions for SNL-UCSB/paper-writing-skill, covering claude.md, what this repo is, installation, architecture and key design decisions.
sapstack abap.instructions.md
Instructions for BoxLogoDev/sapstack, covering abap instructions (github copilot), 기본 규칙 (clean core + hana + k-sox), select 최적화, s/4hana 대응 and 에러 처리.
scientific-writing CLAUDE.md
Instructions for koljaschoepe/scientific-writing, covering scientific writing framework, schnellstart, workflow, befehle and regeln für generierte texte.
sourceright AGENTS.md
AGENTS.md instructions for edithatogo/sourceright, covering sourceright codex instructions, codev - ai agent instructions, project overview, available protocols and file resolution (how codev finds protocols and templates).
avoid-overkill AGENTS.md
AGENTS.md instructions for 6Kmfi6HP/avoid-overkill: This repository packages a reusable Agent Skill and plugin manifests.
academic-writer-skills AGENTS.md
AGENTS.md instructions for muhammad1438/academic-writer-skills, covering academic writer skills — multi-agent workflow guide, skill dependency map, pipeline patterns, pattern 1: full research paper and pattern 2: grant application pipeline.