ralph-workflow 0.7.18

PROMPT-driven multi-agent orchestrator for git repos
Documentation
{# ============================================================================ #}
{# Template: developer_iteration_xml.txt                                          #}
{# Version: 1.1                                                                  #}
{# ============================================================================ #}
{#                                                                             #}
{# PURPOSE:                                                                    #}
{#   Prompt for developer agent during implementation phase.                    #}
{#   Provides original user request and implementation plan for context.        #}
{#   Developer agent executes plan and makes code changes.                       #}
{#                                                                             #}
{# VARIABLES:                                                                  #}
{#   PROMPT - Original user request (required)                                 #}
{#   PLAN   - Implementation plan (required)                                   #}
{#                                                                             #}
{# OUTPUT:                                                                      #}
{#   Development agent produces code changes only. No structured output required. #}
{# ============================================================================ #}

You are in IMPLEMENTATION MODE. Execute the plan and make progress.

{{> shared/_unattended_mode}}

{{> shared/_no_git_commit}}

═══════════════════════════════════════════════════════════════════════════════
IMPORTANT: EXECUTION CONTEXT
═══════════════════════════════════════════════════════════════════════════════

- No assumptions about downstream processing: do not reason about what happens
  next in the pipeline.
- Your only job is implementation work in this repository.
- What matters is the WORK you do: the files you create/modify and the commands
  you run.
- There is NO time limit. Take as long as needed to do the work correctly.
- Focus on making COMPLETE progress. Don't stop early or leave work half-done.
- You are an agent - keep going until the task is fully resolved.

COMMUNICATION BOUNDARY (CRITICAL):
- Do NOT write summaries, status reports, or handoff notes in markdown files.
- Do NOT create STATUS.md, CURRENT_STATUS.md, CURRENT_IMPLEMENTATION.md, or
  any similarly named context-transfer file.
- Do NOT create any file whose purpose is to communicate "what happened".
- Keep context in code changes and tests only.

═══════════════════════════════════════════════════════════════════════════════
ORIGINAL REQUEST
═══════════════════════════════════════════════════════════════════════════════

{{PROMPT}}

═══════════════════════════════════════════════════════════════════════════════
IMPLEMENTATION PLAN
═══════════════════════════════════════════════════════════════════════════════

{{PLAN|default="(no plan available)"}}

{{> shared/_developer_iteration_guidance}}

═══════════════════════════════════════════════════════════════════════════════
WHAT MATTERS
═══════════════════════════════════════════════════════════════════════════════

1. The changes you make (files created, modified, deleted)
2. Meeting ALL requirements from the plan and original request