TTokenySpace
返回 Skills 列表

analyze-project

Rigor Analyze / Rigor Audit read-only skill for deep learning research repositories. Use when the user wants to read and understand a repository, inspect model structure and training or inference entr

0

安装到 Tokeny(自动)

下载 ZIP
安装"analyze-project"技能
技能信息:
- 名称: analyze-project
- 标识: analyze-project
- 描述: Rigor Analyze / Rigor Audit read-only skill for deep learning research repositories. Use when the user wants to read and understand a repository, inspect model structure and training or inference entr
- 版本: 1f652d9
下载地址:
https://www.tokeny.space/api/skills/analyze-project/download
继续

复制上方内容到 Tokeny 客户端并在会话中发送即可自动安装;也可直接 下载 ZIP并拖动到技能窗口安装。

SKILL.md

analyze-project

Use this as the Rigor Analyze / Rigor Audit read-only skill. The installed slug remains analyze-project for compatibility.

Use the shared operating principles in ../../references/agent-operating-principles.md; this skill should guide read-only analysis without constraining the model's project-specific reasoning.

When to apply

  • The user wants to understand a deep learning repository before changing it.
  • The user needs a map of model structure, training entrypoints, inference entrypoints, and config relationships.
  • The user wants conservative suggestions about likely insertion points or suspicious implementation patterns.
  • The user explicitly wants read-only analysis and not heavy execution.

When not to apply

  • When the main task is to execute a failing command or debug a traceback.
  • When the user wants environment setup or asset download only.
  • When the user wants speculative adaptation or broad exploratory patching.
  • When the task is a general literature summary without repository analysis.

Clear boundaries

  • This skill is read-mostly.
  • It may run lightweight static inspection helpers.
  • It does not patch repository code.
  • It does not own final reproduction outputs.
  • It should mark suspicious patterns as heuristics, not confirmed bugs.

Output expectations

  • analysis_outputs/SUMMARY.md
  • analysis_outputs/RISKS.md
  • analysis_outputs/status.json

Notes

Use references/analysis-policy.md and the shared ../../references/research-pitfall-checklist.md.

评论

加载中…