error-doctor

What I do

Diagnose errors systematically using stack trace analysis
Classify error types (SyntaxError, ImportError, TypeError, etc.)
Identify root causes and suggest fixes
Provide actionable remediation steps

When to use me

Use this when you encounter:

Unexpected exceptions or crashes
Error messages you don't understand
Debugging complex failure scenarios
Need systematic approach to error fixing

MCP-First Workflow

Always use MCP servers in this order:

codebase - Search for error handling patterns

python

search_codebase("error handling patterns Python exception", top_k=10)

sequential-thinking - Analyze the error

python

think_step_by_step("Analyze the error stack trace and identify root cause...")

filesystem - view_file the error source

python

read_file("src/module.py", offset=100, limit=50)

git - Check recent changes
python
```
git_diff("HEAD~5..HEAD", path="src/")
```

Error Diagnosis Framework

Error Type	Indicator	Action
SyntaxError	"SyntaxError"	Fix code syntax
ImportError	"ModuleNotFoundError"	Check imports, install deps
TypeError	"expected X got Y"	Fix type mismatch
ValueError	"invalid value"	Validate input
RuntimeError	Generic runtime	Debug logic flow
NetworkError	Connection/timeout	Check service, increase timeout

Common Fixes

Error Pattern	Fix
`NoneType has no attribute`	Add null checks with `get()` or `or`
`list index out of range`	Validate index bounds
`division by zero`	Check denominator before division
`connection refused`	Verify service is running
`timeout expired`	Increase timeout or implement retry

Example Usage

Error: KeyError: 'key' in src/module.py:120

Diagnosis:
1. The code tries to access data["key"] but 'key' doesn't exist
2. Check if the key exists with data.get("key", default_value)
3. Verify the data source provides the expected key

Search AI Tools

Install this agent skill to your Project

Metadata

SKILL.md

What I do

When to use me

MCP-First Workflow

Error Diagnosis Framework

Common Fixes

Example Usage