name	be-better
description	Rewrite user's original prompt by integrating their critique of the agent's failed response.
argument-hint	What did the agent do wrong? What details should be added to the initial request?
disable-model-invocation	true
stage	drafted

Goal: Synthesize the original prompt and current critique into a single, flawless replacement prompt.

Rules:

Context Fusion: Analyze original request + agent's inadequate response + user's current critique. Merge them into ONE comprehensive prompt.
Prevent Failure: Explicitly define constraints, context, or negative prompts to block the mistakes the agent just made.
Actionable & Unambiguous: Replace vague terms from the original prompt with precise technical directives.
Language Match: Output in the exact language of the user's original request.
Strict Output: Return ONLY the raw text of the improved prompt. NEVER use conversational filler, explanations, or wrap the entire output in markdown code blocks. It must be 100% copy-paste ready.

name	be-better
description	Rewrite user's original prompt by integrating their critique of the agent's failed response.
argument-hint	What did the agent do wrong? What details should be added to the initial request?
disable-model-invocation	true
stage	drafted

Goal: Synthesize the original prompt and current critique into a single, flawless replacement prompt.

Context Fusion: Analyze original request + agent's inadequate response + user's current critique. Merge them into ONE comprehensive prompt.
Prevent Failure: Explicitly define constraints, context, or negative prompts to block the mistakes the agent just made.
Actionable & Unambiguous: Replace vague terms from the original prompt with precise technical directives.
Language Match: Output in the exact language of the user's original request.
Strict Output: Return ONLY the raw text of the improved prompt. NEVER use conversational filler, explanations, or wrap the entire output in markdown code blocks. It must be 100% copy-paste ready.

be-better