Back to all prompts
ClaudeData Analysis
FreeAgent State Classification
Classify background agent states with examples and JSON outputs, optimized for Claude AI. Useful for data analysis and automation.
P
Piebald-AI4.5
The Prompt
<!--
name: 'Data: Background agent state classification examples'
description: Example assistant-message tails and JSON outputs for classifying background agent state, tempo, needs, and result
ccVersion: 2.1.119
-->
EXAMPLES (message → classification):
"Reading config files to understand the setup."
→ {"state":"working","detail":"reading config files","tempo":"active","output":{}}
"I found the bug in auth.ts:42. Want me to fix it or just report?"
→ {"state":"blocked","detail":"found bug, awaiting direction","tempo":"blocked","needs":"Want me to fix it or just report?","output":{}}
"PR opened: https://github.com/acme/repo/pull/123\nresult: fixed auth race in auth.ts, PR #123"
→ {"state":"done","detail":"opened PR #123","tempo":"idle","output":{"result":"fixed auth race in auth.ts, PR #123"}}
"I can't proceed — the repo requires GITHUB_TOKEN and it's not set."
→ {"state":"blocked","detail":"missing GITHUB_TOKEN","tempo":"blocked","needs":"set GITHUB_TOKEN env var","output":{}}
"Can't run the tests — needs the openapi.yaml file which isn't in this checkout. Stopping here."
→ {"state":"blocked","detail":"missing openapi.yaml","tempo":"blocked","needs":"provide config/openapi.yaml","output":{}}
("stopping" + names a specific missing resource → blocked, not failed)
"The build is broken on main and I can't reproduce locally. Giving up."
→ {"state":"failed","detail":"cannot reproduce build failure","tempo":"idle","output":{}}
(no specific resource would unblock; exhausted approaches → failed)
"Tests pass. Let me know if you want me to also update the docs."
→ {"state":"done","detail":"tests pass","tempo":"idle","output":{"result":"tests pass"}}
(offer of optional extra work ≠ blocked; the ask is satisfied)
"Waiting for CI to finish (~8 min)."
→ {"state":"working","detail":"waiting for CI","tempo":"idle","output":{}}
"API Error: 401 Invalid API key · Please run /login"
→ {"state":"blocked","detail":"authentication failed","tempo":"blocked","needs":"run /login","output":{}}
#claude#data#analysis#agent#state#classification
Source: Piebald-AI/claude-code-system-prompts by Piebald-AI · License: MIT
Related Prompts
Gemini$1.00
Data Visualization Expert
Get expert data visualization recommendations with chart selection, design, and Python code using Gemini AI. Create effe...
Data Analysisdatavisualization
by DataVizPro
4.6
GeneralFree
Ultra-Wise Content Classifier
Rates content with single-word labels and S-D tier scale combined with a 1-100 numeric score, ideal for data analysis wi...
Data Analysisratingclassification
by Daniel Miessler
4.9
GeneralFree
Analyze Hidden Messages
Expert AI analyzes hidden messages in content, providing cynical, balanced, and favorable perspectives, ideal for use wi...
Data Analysispolitical-analysispropaganda
by Daniel Miessler
4.6