agentic-evidence-lab good first issues (Python)

Threat-model short-lived credential brokerage for external submissions 2 months ago

help wanted research needs-design difficulty:flagship security-design

1

Tests exact agent interventions—skills, prompts, models, tools, and workflows—and publishes what changed, what failed, and what decision follows.

Python

#agent-evaluation#agent-skills#agentic-ai#ai-agents#benchmarking#codex#coding-agents#evals#llm-agents#llm-evaluation#reproducibility#software-engineering

Prepare a role-separated reproduction of one eligible AEL decision 2 months ago

help wanted research replication needs-design difficulty:flagship

kizz-tech/agentic-evidence-lab

1

Tests exact agent interventions—skills, prompts, models, tools, and workflows—and publishes what changed, what failed, and what decision follows.

Python

#agent-evaluation#agent-skills#agentic-ai#ai-agents#benchmarking#codex#coding-agents#evals#llm-agents#llm-evaluation#reproducibility#software-engineering

Generate embeddable badges from validated AEL decision cards 2 months ago

enhancement help wanted community needs-design difficulty:advanced

kizz-tech/agentic-evidence-lab

1

Tests exact agent interventions—skills, prompts, models, tools, and workflows—and publishes what changed, what failed, and what decision follows.

Python

#agent-evaluation#agent-skills#agentic-ai#ai-agents#benchmarking#codex#coding-agents#evals#llm-agents#llm-evaluation#reproducibility#software-engineering

Design an offline run-import boundary for Claude Code and OpenCode 2 months ago

help wanted runner-adapter needs-design difficulty:advanced

kizz-tech/agentic-evidence-lab

1

Tests exact agent interventions—skills, prompts, models, tools, and workflows—and publishes what changed, what failed, and what decision follows.

Python

#agent-evaluation#agent-skills#agentic-ai#ai-agents#benchmarking#codex#coding-agents#evals#llm-agents#llm-evaluation#reproducibility#software-engineering

Make release verification portable across Linux, macOS, and Windows 2 months ago

help wanted python needs-design difficulty:advanced

kizz-tech/agentic-evidence-lab

1

Tests exact agent interventions—skills, prompts, models, tools, and workflows—and publishes what changed, what failed, and what decision follows.

Python

#agent-evaluation#agent-skills#agentic-ai#ai-agents#benchmarking#codex#coding-agents#evals#llm-agents#llm-evaluation#reproducibility#software-engineering

Add a public four-state evaluator fixture for task-pack authors 2 months ago

good first issue help wanted task-pack

kizz-tech/agentic-evidence-lab

1

Tests exact agent interventions—skills, prompts, models, tools, and workflows—and publishes what changed, what failed, and what decision follows.

Python

#agent-evaluation#agent-skills#agentic-ai#ai-agents#benchmarking#codex#coding-agents#evals#llm-agents#llm-evaluation#reproducibility#software-engineering

Create a one-command "Hello Study" walkthrough 2 months ago

documentation good first issue community

kizz-tech/agentic-evidence-lab

1

Tests exact agent interventions—skills, prompts, models, tools, and workflows—and publishes what changed, what failed, and what decision follows.

Python

#agent-evaluation#agent-skills#agentic-ai#ai-agents#benchmarking#codex#coding-agents#evals#llm-agents#llm-evaluation#reproducibility#software-engineering

kizz-tech/agentic-evidence-lab

Open Issues Need Help