# Explain

> Explain a file, function, or recent diff in plain language — what it does, how it flows, and why. Use when the user asks to explain code, says "what does this do", "walk me through this file", "help me understand this", "explain the diff", or is onboarding to unfamiliar code.

- **Type:** Skill
- **Install:** `agentstack add skill-pradnyeshp-claude-skills-explain`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [Pradnyeshp](https://agentstack.voostack.com/s/pradnyeshp)
- **Installs:** 0
- **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [Pradnyeshp](https://github.com/Pradnyeshp)
- **Source:** https://github.com/Pradnyeshp/Claude-Skills/tree/main/.claude/skills/explain

## Install

```sh
agentstack add skill-pradnyeshp-claude-skills-explain
```

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

## About

## Recent diff (used when the user asks to explain "the diff" or "the changes")

!`git diff HEAD --stat 2>/dev/null || echo "No git changes detected"`

## Instructions

Explain the target the user named in `$ARGUMENTS` (a file path, a function/symbol name, or "diff"/"changes" to explain recent edits). If no target was given, ask what they'd like explained.

### Approach

1. **Read before explaining.** Open the relevant file(s). For a symbol, use Grep to find its definition and key call sites. For "diff", run `git diff HEAD` and explain those changes.
2. Lead with a **one-sentence summary** of what the code does and why it exists.
3. Then give a **structured walkthrough**:
   - Inputs and outputs (parameters, return values, side effects)
   - The main control flow, step by step
   - Key data structures or algorithms involved
   - External dependencies it touches (DB, network, files, other modules)
4. Call out anything **non-obvious or surprising** — implicit assumptions, edge-case handling, performance characteristics, or footguns.
5. If the code looks buggy or fragile, note it briefly at the end, but don't turn this into a full review.

### Rules

1. Match the explanation depth to the audience — assume a competent engineer new to *this* codebase, not new to programming.
2. Reference real symbols and `file:line` locations; don't speak in generalities.
3. Use short paragraphs and bullet points. Include a tiny annotated snippet only when it clarifies a tricky part.
4. Do not modify any files — this skill is read-only.

## Source & license

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

- **Author:** [Pradnyeshp](https://github.com/Pradnyeshp)
- **Source:** [Pradnyeshp/Claude-Skills](https://github.com/Pradnyeshp/Claude-Skills)
- **License:** MIT

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-pradnyeshp-claude-skills-explain
- Seller: https://agentstack.voostack.com/s/pradnyeshp
- 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%.