# Academic Repo Analyzer

> Quick-understanding doc for ML/DL repositories — task type, stack, architecture, and figure-worthy innovations. Use when the user wants repo analysis, 仓库分析, or to understand a codebase before figure planning.

- **Type:** Skill
- **Install:** `agentstack add skill-azhi-ss-academic-figure-skills-academic-repo-analyzer`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [Azhi-ss](https://agentstack.voostack.com/s/azhi-ss)
- **Installs:** 0
- **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [Azhi-ss](https://github.com/Azhi-ss)
- **Source:** https://github.com/Azhi-ss/academic-figure-skills/tree/main/academic-repo-analyzer
- **Website:** https://www.skills.sh/azhi-ss/academic-figure-skills

## Install

```sh
agentstack add skill-azhi-ss-academic-figure-skills-academic-repo-analyzer
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# Academic Repo Analyzer

Produce a **仓库快速理解文档** for downstream figure planning.

Keywords: → `keywords.md`  
Missing info: → `../docs/missing-info-policy.md`

## Input Contract

- Prefer: repo path, README, deps, entry scripts, model files, configs
- Minimum: any one of README / entry script / model file
- Missing: partial analysis with 推断 / 待确认

## Output Contract — Quick Understanding Doc

- overview (name, task, framework, architecture one-liner)
- completeness block
- stack details
- model / algorithm notes
- train / inference flow (or “evidence insufficient”)
- figure suggestions for paper-analyzer

## Steps

### Step 1: Scan structure

Locate README, dependency files, entry scripts (`train|main|eval|inference`), `configs/`, `models|networks|src/`, data loaders.

Done when: tree of key paths exists and each must-read class is read **or** marked missing.

### Step 2: Task + stack

Use `keywords.md`. Classify task type and framework from imports, deps, and paths.

Done when: task type + primary framework are stated with file evidence.

### Step 3: Architecture + algorithms

From model files: backbone family, key modules, losses, training tricks. Prefer evidence over naming guesses.

Done when: architecture summary cites concrete classes/files, or is marked 推断.

### Step 4: Emit quick-understanding doc

```markdown
# 仓库快速理解文档
## 仓库概览
| 项目 | 内容 |
| 仓库名称 / 任务类型 / 核心框架 / 主要架构 / 一句话描述 | ... |
## 信息完整度说明
## 技术栈详情
## 模型架构分析
## 工作流程
## 配图建议（→ paper-analyzer）
```

Done when: Output Contract fields are filled; figure suggestions list concrete types (framework / arch / module / …).

## Sparse-input cases

| gap | action |
|-----|--------|
| no README | infer from code; label as structure-inferred |
| no entry scripts | module-level understanding only |
| no model files | stack/task only; soft architecture language |
| huge repo | sample top-level + 3–5 core files; mark limited sample |
| almost nothing | pre-analysis + minimum materials list (README → deps → entry → model → config) |

## Tooling cues

Prefer the environment’s file/search tools. Typical digs: dependency files, `class.*Model|Network|Transformer`, `loss|criterion`, model package entrypoints.

## Stop

Stop when the quick-understanding doc is delivered. Suggest paper-analyzer only if the user wants figure planning next.

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [Azhi-ss](https://github.com/Azhi-ss)
- **Source:** [Azhi-ss/academic-figure-skills](https://github.com/Azhi-ss/academic-figure-skills)
- **License:** MIT
- **Homepage:** https://www.skills.sh/azhi-ss/academic-figure-skills

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-azhi-ss-academic-figure-skills-academic-repo-analyzer
- Seller: https://agentstack.voostack.com/s/azhi-ss
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.