Analyze Weekly Agent Logs
Review a week's conversation logs for resolution rates, failure points, repeated requests, and prioritized prompt improvements.
Prompt template
- Format
- Text Parameterized
- Inputs
- dialogue_logs (text) · Required
- Outputs
- Weekly_log_analysis (text)
Here are the dialogue logs from the past week:
${dialogue-logs}
Count how many conversations were resolved without human assistance, identify the questions the agent struggled with, and find which requests appeared more than 3 times. Suggest 3 prompt changes for the coming week, ranked by impact. Include only findings that are visible in the logs.
Replace the ${…} placeholders with your values
before using this prompt. See parameters
and defaults below
Parameters
Replace each placeholder with your own value, or use its default when one is provided
${dialogue-logs}
dialogue-logs
Required
What you'll create
A report counts conversations resolved without human help, identifies difficult questions and flags requests occurring more than three times. It ranks three prompt improvements using only visible log evidence.
Frequently asked questions
What is the repetition threshold?
More than three appearances.
Can the report infer unseen problems?
No. Findings must be visible in the provided logs.
Comments
Tried this prompt? Share what you made, compare results, or ask a question