# Debug

> Use when encountering any bug, test failure, or unexpected behavior. Requires root cause investigation before proposing fixes. No guessing.

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

## Install

```sh
agentstack add skill-sandsower-beislid-debug
```

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

## About

# Debug Method

Find the root cause before touching code. No quick fixes, no guessing, no shotgun debugging.

Do NOT propose or implement a fix until you have identified the root cause with evidence. "It's probably X" is not a root cause.

## Phases

### 1. Reproduce
- Read the error message carefully — the full stack trace, not just the last line
- Reproduce the failure consistently
- If you can't reproduce it, you don't understand it yet

### 2. Investigate
- Check recent changes (`git log`, `git diff`) that could have introduced the issue
- Find a working reference — when did this last work? What's different?
- Trace the data flow from input to error point
- In multi-component systems, isolate which component fails first

### 3. Hypothesize
- Form a single hypothesis based on evidence gathered
- Test it minimally — the smallest change that proves or disproves the hypothesis
- If disproved, go back to step 2 with new information
- If you're stuck after 3 hypotheses, say "I don't understand this yet" and surface to the user

### 4. Fix
- Write a failing test that captures the bug
- Implement the fix — one change at a time
- Verify the test goes from red to green
- Run the full suite to check for regressions

## Red Flags

Stop yourself if you're thinking:
- "Let me just try changing X" — that's guessing
- "Quick fix for now" — that's deferring understanding
- "Multiple changes at once" — that's making it harder to know what worked
- "I don't fully understand but this should work" — go back to investigate

## Source & license

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

- **Author:** [sandsower](https://github.com/sandsower)
- **Source:** [sandsower/beislid](https://github.com/sandsower/beislid)
- **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-sandsower-beislid-debug
- Seller: https://agentstack.voostack.com/s/sandsower
- 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%.