Test-driven harness engineering for Python agents: own the loop, capture failures, and gate every change.

6 stars 4 forks 6 watchers Python Apache License 2.0
agent-evals agent-harness ai-agents anthropic asyncio function-calling harness-engineering hooks llm llm-agent mcp observability ollama openai provenance pytest python regression-testing testing tool-calling
17 Open Issues Need Help Last updated: Jul 16, 2026

Open Issues Need Help

View All on GitHub
documentation help wanted good first issue provider launch-ready

Test-driven harness engineering for Python agents: own the loop, capture failures, and gate every change.

Python
#agent-evals#agent-harness#ai-agents#anthropic#asyncio#function-calling#harness-engineering#hooks#llm#llm-agent#mcp#observability#ollama#openai#provenance#pytest#python#regression-testing#testing#tool-calling
documentation help wanted good first issue launch-ready

Test-driven harness engineering for Python agents: own the loop, capture failures, and gate every change.

Python
#agent-evals#agent-harness#ai-agents#anthropic#asyncio#function-calling#harness-engineering#hooks#llm#llm-agent#mcp#observability#ollama#openai#provenance#pytest#python#regression-testing#testing#tool-calling
enhancement help wanted evals provenance launch-ready

Test-driven harness engineering for Python agents: own the loop, capture failures, and gate every change.

Python
#agent-evals#agent-harness#ai-agents#anthropic#asyncio#function-calling#harness-engineering#hooks#llm#llm-agent#mcp#observability#ollama#openai#provenance#pytest#python#regression-testing#testing#tool-calling
help wanted evals provenance api-design launch-ready

Test-driven harness engineering for Python agents: own the loop, capture failures, and gate every change.

Python
#agent-evals#agent-harness#ai-agents#anthropic#asyncio#function-calling#harness-engineering#hooks#llm#llm-agent#mcp#observability#ollama#openai#provenance#pytest#python#regression-testing#testing#tool-calling
enhancement help wanted evals launch-ready

Test-driven harness engineering for Python agents: own the loop, capture failures, and gate every change.

Python
#agent-evals#agent-harness#ai-agents#anthropic#asyncio#function-calling#harness-engineering#hooks#llm#llm-agent#mcp#observability#ollama#openai#provenance#pytest#python#regression-testing#testing#tool-calling
documentation help wanted good first issue provenance launch-ready

Test-driven harness engineering for Python agents: own the loop, capture failures, and gate every change.

Python
#agent-evals#agent-harness#ai-agents#anthropic#asyncio#function-calling#harness-engineering#hooks#llm#llm-agent#mcp#observability#ollama#openai#provenance#pytest#python#regression-testing#testing#tool-calling
documentation help wanted good first issue eval-integrity launch-ready

Test-driven harness engineering for Python agents: own the loop, capture failures, and gate every change.

Python
#agent-evals#agent-harness#ai-agents#anthropic#asyncio#function-calling#harness-engineering#hooks#llm#llm-agent#mcp#observability#ollama#openai#provenance#pytest#python#regression-testing#testing#tool-calling
enhancement help wanted provider launch-ready

Test-driven harness engineering for Python agents: own the loop, capture failures, and gate every change.

Python
#agent-evals#agent-harness#ai-agents#anthropic#asyncio#function-calling#harness-engineering#hooks#llm#llm-agent#mcp#observability#ollama#openai#provenance#pytest#python#regression-testing#testing#tool-calling
enhancement help wanted provider launch-ready

Test-driven harness engineering for Python agents: own the loop, capture failures, and gate every change.

Python
#agent-evals#agent-harness#ai-agents#anthropic#asyncio#function-calling#harness-engineering#hooks#llm#llm-agent#mcp#observability#ollama#openai#provenance#pytest#python#regression-testing#testing#tool-calling

Test-driven harness engineering for Python agents: own the loop, capture failures, and gate every change.

Python
#agent-evals#agent-harness#ai-agents#anthropic#asyncio#function-calling#harness-engineering#hooks#llm#llm-agent#mcp#observability#ollama#openai#provenance#pytest#python#regression-testing#testing#tool-calling
documentation enhancement help wanted

Test-driven harness engineering for Python agents: own the loop, capture failures, and gate every change.

Python
#agent-evals#agent-harness#ai-agents#anthropic#asyncio#function-calling#harness-engineering#hooks#llm#llm-agent#mcp#observability#ollama#openai#provenance#pytest#python#regression-testing#testing#tool-calling
enhancement help wanted good first issue

Test-driven harness engineering for Python agents: own the loop, capture failures, and gate every change.

Python
#agent-evals#agent-harness#ai-agents#anthropic#asyncio#function-calling#harness-engineering#hooks#llm#llm-agent#mcp#observability#ollama#openai#provenance#pytest#python#regression-testing#testing#tool-calling
documentation help wanted good first issue

Test-driven harness engineering for Python agents: own the loop, capture failures, and gate every change.

Python
#agent-evals#agent-harness#ai-agents#anthropic#asyncio#function-calling#harness-engineering#hooks#llm#llm-agent#mcp#observability#ollama#openai#provenance#pytest#python#regression-testing#testing#tool-calling
enhancement help wanted good first issue

Test-driven harness engineering for Python agents: own the loop, capture failures, and gate every change.

Python
#agent-evals#agent-harness#ai-agents#anthropic#asyncio#function-calling#harness-engineering#hooks#llm#llm-agent#mcp#observability#ollama#openai#provenance#pytest#python#regression-testing#testing#tool-calling
enhancement help wanted good first issue

Test-driven harness engineering for Python agents: own the loop, capture failures, and gate every change.

Python
#agent-evals#agent-harness#ai-agents#anthropic#asyncio#function-calling#harness-engineering#hooks#llm#llm-agent#mcp#observability#ollama#openai#provenance#pytest#python#regression-testing#testing#tool-calling
enhancement help wanted good first issue

Test-driven harness engineering for Python agents: own the loop, capture failures, and gate every change.

Python
#agent-evals#agent-harness#ai-agents#anthropic#asyncio#function-calling#harness-engineering#hooks#llm#llm-agent#mcp#observability#ollama#openai#provenance#pytest#python#regression-testing#testing#tool-calling
documentation help wanted good first issue

Test-driven harness engineering for Python agents: own the loop, capture failures, and gate every change.

Python
#agent-evals#agent-harness#ai-agents#anthropic#asyncio#function-calling#harness-engineering#hooks#llm#llm-agent#mcp#observability#ollama#openai#provenance#pytest#python#regression-testing#testing#tool-calling