Installs into .claude/skills of the current project.
Are you the author of 2019 Retellai Known Pitfalls 838b5526?
Add the live security badge to your README. It updates with every re-scan.
[](https://www.skillsdirectory.com/skills/tools-only-2019-retellai-known-pitfalls-838b5526)
# retellai-known-pitfalls
> Identify and avoid 10 common anti-patterns including blocking calls, poor error handling, and voice misuse
## Directory Structure
```
retellai-known-pitfalls/
├── SKILL.md
└── examples/
└── example.py
```
## File Descriptions
| File | Type | Purpose |
|------|------|---------|
| SKILL.md | Markdown | Core skill instructions for avoiding common anti-patterns |
| examples/example.py | Python | Example good patterns vs anti-patterns comparison |
## Summary
**Category:** advanced
**Target Audience:** All Retell AI developers, Code reviewers, Tech leads
**Trigger Phrases:** `retell mistakes`, `retell anti-patterns`, `retell pitfalls`, `retell what not to do`
### What This Skill Does
This skill identifies and helps developers avoid the 10 most common Retell AI anti-patterns. It covers blocking API calls in voice handlers (causing latency), ignoring error responses, hardcoding API keys, mismatched voice/persona, overly complex conversation flows, missing webhook signature verification, ignoring rate limits, synchronous transcript processing, missing health checks, and poor logging practices.
### Technical Success Criteria
- Anti-patterns identified in existing code
- Fixes prioritized by impact
- Prevention measures in place (linting, code review)
- Team educated on common pitfalls
### Business Success Criteria
- Reduced production incidents from known issues
- Improved code quality across team
- Eliminate 100% of common pitfalls in new code
## Related Skills
- retellai-reference-architecture - Correct patterns to follow
- retellai-security-basics - Security-related pitfalls
- retellai-common-errors - Error resolution