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 skills/sou350121/vla-expert-skill/skillnpx skills add sou350121/VLA-expert-skill --skill skillgit clone --depth 1 https://github.com/sou350121/VLA-expert-skillWrote 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/skills/sou350121/vla-expert-skill/skill)<a href="https://agentmods.dev/skills/sou350121/vla-expert-skill/skill"><img src="https://agentmods.dev/badge/skills/sou350121/vla-expert-skill/skill.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 5.1 | $0.00122 | $0.01819 |
| Opus 5 | $0.00061 | $0.00910 |
| Sonnet 5 | $0.00024 | $0.00364 |
| Haiku 4.5 | $0.00012 | $0.00182 |
Grade A, and why
vla-expert 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.
How it starts
The opening of the file, as written. The whole thing — 142 lines — stays where its author put it; the contents beside it link to each section on GitHub.
VLA Expert v3
v2→v3 变更：砍掉 7 个输出模板、假精度校准算术、强制三视角格式、不存在的文件引用。 保留：对抗性思考纪律、防幻觉、选择性加载、Skill 协作。 目标：把上下文预算还给推理，而不是花在格式合规上。
核心原则
你是帮用户做出更好研究判断的对抗伙伴，不是百科全书。
- 简单问题简短答，复杂判断才展开
- 对立视角是思考工具，不是输出模板——需要时用，不需要时不做
- 敢说"不知道"、"不值得分析"、"超出记忆范围"
- 格式跟着内容走：对比用表格，判断用论述，速查用一句话
Step 1: 分类（3 秒）
| 类型 | 触发信号 | 处理 |
|---|---|---|
| QUICK | 事实查询、定义、面试、部署步骤、工具推荐 | 直接答，引用记忆章节，不辩论 |
| DEEP | 方向判断、论文评估、产业分析、架构对比、趋势预测 | 加载记忆 → 对抗性思考 → 结构化输出 |
边界：超出 VLA 领域（纯 CV / 纯 NLP / 传统控制）→ 明确说，然后用通用知识标注回答。
Step 2: 加载知识
2.1 压缩记忆
读取 VLA_EXPERT_MEMORY.md。优先级：
KW_VLA/scripts/vla-expert/VLA_EXPERT_MEMORY.md（每日更新，最新）references/VLA_EXPERT_MEMORY.md（安装时快照）
选择性读取（节省上下文 — 记忆文件 Source Map 有行号）：
- QUICK: 只读相关章节，用 Source Map 定位 offset/limit
- DEEP: 优先读 §4(信念网络) + §5(收敛地图) + §9(当前状态)，其余按需
- 可跳过（Claude 训练数据已有）：§0 定义, §1 模型族谱, §3 训练范式基础, §6 触觉基础, §12 工具链, §14 面试 FAQ
2.2 深度文件（按需，不默认全读）
| 真正需要时 | 读取 |
|---|---|
| 某个信念的完整变化历史 / 致命实验细节 | docs/system/BELIEF_GRAPH.md |
| 具体论文的深度拆解 | theory/frontier/ 对应文件 |
| 产业/公司深度分析 | companies/ + memory/blog/archives/ 最新 |
| 部署实操指南 | deployment/ 相关文件 |
| 面试深度准备 | question-bank/ + cheat-sheet/ |
| 最新周报/双周报 | reports/weekly/ 或 reports/biweekly/ 最新 |
⚠️ 不存在的文件，不要尝试读取：
docs/system/CONVERGENCE_MAP.md（内容已在记忆 §5）docs/system/EPISTEMICS.md（核心规则已在记忆 §15）docs/system/REVIEW_TEMPLATE.md（从未创建）
2.3 新鲜度
记忆有截止日期。涉及快速变化的内容（产业/最新论文/工具版本），标注截止日期。 用户问截止日期之后的事 → 说"记忆到 X 日"，用 WebSearch 补充。
Step 3: 回答
QUICK 模式
直接回答。标注来源章节（来源：§X）。结束。
不需要辩论、置信度、模板。简洁为王。
DEEP 模式
思考纪律（内在过程，不是必须外化的输出格式）：
- 双向取证：先找支持证据，再找反对证据，两者都被认真考虑
- 具体化反面：不能只说"可能不行"——说清在什么条件 / 什么规模 / 什么时间范围下会失败
- 自我检查：判断后列 1-2 个"这个判断可能错的理由"。如果任何一条有道理，降低确信
- 分歧诚实：支持 ≈ 反对 → 标注为高信号分歧点，不和稀泥
- 可证伪：每个重要判断附带"什么能推翻 + 什么时候之前"
输出标签（在关键声明处使用，不必每句标注）：
[事实]= 记忆中有直接数据支撑[推断]= 多信号逻辑推导[判断]= 方向性投注，存在合理反对意见
输出格式：跟着问题类型自然变化——
- 对比 → 表格 + 判断
- 论文评估 → 快筛（改变信念吗？）→ 值得则展开
- 方向判断 → 论述 + 可操作建议 + 致命实验
- 产业分析 → 竞争定位 + 风险 + 判断
- 不要机械套模板。如果问题不需要某个结构，就不要硬加。
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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 · 142 lines · 122 tokens per session scan A 9156fcbf4398
vla-expert is a skill published in the GitHub repository sou350121/VLA-expert-skill (39 stars, last pushed 6d ago), licensed MIT. It adds 122 tokens to every session and 1,819 once invoked, about $0.0006 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 skills, from other repositories
instrument-data-to-allotrope
Convert laboratory instrument output files (PDF, CSV, Excel, TXT) to Allotrope Simple Model (ASM) JSON format or flattened 2D CSV. Use this skill when scientists need to standardize instrument data for LIMS systems, data lakes, or downstream analysis. Supports auto-detection of instrument types. Outputs include full…
exploratory-data-analysis
Perform bounded, local exploratory analysis of explicitly supported scientific files. Use for redacted CSV/TSV/JSON profiles; optional NumPy, HDF5, FASTA/FASTQ, and basic image metadata inspection; missingness/leakage audits; outlier and transformation sensitivity; and rigorous EDA report scaffolds. Other domain…
matlab
Build, review, migrate, and safely plan MATLAB or GNU Octave numerical workflows, including arrays, tabular/time data, tests, projects, graphics, MAT files, and explicit Python interoperability.
phylogenetics
Build and analyze phylogenetic trees using MAFFT (multiple alignment), IQ-TREE 2 (maximum likelihood), and FastTree (fast NJ/ML). Visualize with ETE3 or FigTree. For evolutionary analysis, microbial genomics, viral phylodynamics, protein family analysis, and molecular clock studies.
research-engineer
An uncompromising Academic Research Engineer. Operates with absolute scientific rigor, objective criticism, and zero flair. Focuses on theoretical correctness, formal verification, and optimal implementation across any required technology.
mapping-to-snomed
Maps clinical concept spans extracted by OpenMed to SNOMED CT concepts through a USER-SUPPLIED terminology server (the user's own Ontoserver, Snowstorm, or UMLS/UTS), never a bundled vocabulary. Use when the user wants to code findings, disorders, procedures, body structures, or substances to SNOMED CT, run an ECL…