← 返回 Skills 市场
101
总下载
0
收藏
1
当前安装
1
版本数
在 OpenClaw 中安装
/install methodology-extractor
功能描述
Batch extraction of experimental methods from multiple papers for protocol.
安全使用建议
This skill is coherent in being a local Python utility but is suspicious because its documentation promises file-based batch processing and a CONFIG block that the shipped script does not implement. Before installing or running: (1) inspect scripts/main.py yourself — it's short and currently only supports a built-in demo; it does not read --papers or write outputs; (2) do not run code from unknown sources on sensitive hosts — test in a sandbox or disposable environment; (3) ask the publisher for a corrected implementation or for explicit instructions on how to provide input files and where outputs are written; (4) if you need the advertised functionality, request that the author add safe input validation, explicit CLI handling for input/output paths, and unit tests demonstrating file I/O; (5) because there are no credentials or network calls in the current code, the immediate risk of exfiltration is low, but the documentation/code mismatch is a red flag — treat the package as incomplete until fixed.
能力评估
Purpose & Capability
The skill claims 'batch extraction of experimental methods from multiple papers' and documents file input/output and a CONFIG block, but scripts/main.py only contains a tiny demo and does not implement reading a --papers JSON file, writing an output file, or any CONFIG handling. The delivered capability is much smaller than advertised.
Instruction Scope
SKILL.md instructs the agent/operator to validate inputs, edit an in-file CONFIG block, run the script with inputs, and produce output artifacts. The script offers only a --demo mode and otherwise prints a usage hint; it does not parse files, perform filesystem I/O, or accept parameters for processing. That mismatch gives the agent vague authority to 'edit' files and run code without a concrete, implemented workflow.
Install Mechanism
There is no install spec (instruction-only with a packaged script). This is low-risk from an install/extraction perspective because nothing is downloaded or installed during pack installation.
Credentials
The skill declares no required environment variables, credentials, or config paths and the script does not access environment secrets. No disproportionate credential access is requested.
Persistence & Privilege
The skill is not always-enabled and uses the normal invocation model. It does not modify other skills or system-wide settings and does not request persistent privileges.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install methodology-extractor - 安装完成后，直接呼叫该 Skill 的名称或使用
/methodology-extractor触发 - 根据 Skill 的参数说明提供必要输入，即可获得结构化输出
版本历史
v1.0.0
Initial release of methodology-extractor: batch extraction and comparison of experimental methods from multiple papers.
- Extracts and compares experimental protocols across multiple research papers for evidence insight tasks.
- Provides a scope-focused workflow with clear assumptions, reproducible output, and documented fallback/error handling.
- Includes executable script (scripts/main.py) and reference material for guidance.
- Highlights input validation, output discipline, risk assessment, and security checklist.
- Sample outputs include comparison tables of protocols, parameter variations, and best practice recommendations.
元数据
常见问题
Methodology Extractor 是什么？
Batch extraction of experimental methods from multiple papers for protocol. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 101 次。
如何安装 Methodology Extractor？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install methodology-extractor」即可一键安装，无需额外配置。
Methodology Extractor 是免费的吗？
是的，Methodology Extractor 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Methodology Extractor 支持哪些平台？
Methodology Extractor 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。
谁开发了 Methodology Extractor？
由 AIpoch（@aipoch-ai）开发并维护，当前版本 v1.0.0。
推荐 Skills