Review a Week of Agent Logs
Measure agent outcomes, find recurring failure points and requests, then recommend three prioritized prompt changes grounded in the logs.
Prompt template
- Format
- Text Parameterized
- Inputs
- Logs (text) · Required
- Outputs
- Weekly_log_analysis (text)
Here are the dialogue logs for the week:
${logs}
Count how many conversations were closed without a person. Identify the questions on which the agent failed and the requests that recurred more than 3 times. Recommend 3 changes to the prompt for next week, ordered by impact. Include only findings supported by 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
${logs}
logs
Required
What you'll create
The analysis counts conversations closed without a person and identifies supported failure themes. Three prioritized prompt changes translate the log findings into next-week actions.
Frequently asked questions
What recurrence threshold is specified?
Requests occurring more than three times.
Can findings use unsupported guesses?
No. Only evidence in the logs is allowed.
Comments
Tried this prompt? Share what you made, compare results, or ask a question