# Execute PRP

Implement a feature from its PRP spec file.

## PRP File: $ARGUMENTS

## Execution Process

1. **Load Context**
   - Read the PRP file
   - Read `CLAUDE.md` for conventions
   - Read any referenced specs or models
   - Ensure you understand all interfaces and data contracts

2. **Plan (ULTRATHINK)**
   - Break the PRP into ordered tasks using TodoWrite
   - Identify which files to create/modify
   - Check existing code for patterns to follow
   - If anything is ambiguous, check the spec again before guessing

3. **Execute**
   - Implement each task in order
   - Mark tasks complete as you go
   - After each file, run `ruff check` and `mypy` on it
   - Write tests alongside implementation, not after

4. **Validate**
   - Run all validation commands from the PRP's Success Criteria
   - Fix any failures — do not skip
   - Re-run until all pass

5. **Final Check**
   - Re-read the PRP
   - Verify every Success Criteria checkbox
   - Run the full test suite one last time
   - Report: what was built, what works, any deviations from spec

## Rules
- Never skip type hints
- Never hardcode API keys
- Always use shared models from `backend/src/models/`
- Paper trading is DEFAULT — never enable live trading in code
- Log before every external API call

Maintain Execute Prp?

Let people know it's listed here — add the badge (live metrics, light/dark aware) or a plain link to your README or docs.

[Execute Prp on getagentictools](https://getagentictools.com/loops/pmxlr8-execute-prp?ref=badge)

npx agentictools info loops/pmxlr8-execute-prp

The second line is the CLI lookup for this page — handy in READMEs and docs.

Maintain Execute Prp?

Related loops