agent-harness-skills good first issues (Python)

failure mode wanted: prompt-cache invalidation in a long agent loop about 2 months ago

help wanted failure-mode

1

IDE skills for scaffolding AI agent harnesses: ReAct, plan-and-execute, self-evolving, reflexion, multi-agent, code agent, RAG memory

Python

#agent-framework#agent-harness#agents#agents-md#ai-agents#claude-code#good-first-issue#harness#llm-agents#multi-agent#opencode#rag#react#reflexion#self-improving#skills

failure mode wanted: timeout or cancellation part-way through a tool call about 2 months ago

help wanted failure-mode

ArttuAn/agent-harness-skills

1

IDE skills for scaffolding AI agent harnesses: ReAct, plan-and-execute, self-evolving, reflexion, multi-agent, code agent, RAG memory

Python

#agent-framework#agent-harness#agents#agents-md#ai-agents#claude-code#good-first-issue#harness#llm-agents#multi-agent#opencode#rag#react#reflexion#self-improving#skills

failure mode wanted: structured output / JSON mode colliding with tool calling about 2 months ago

help wanted failure-mode

ArttuAn/agent-harness-skills

1

IDE skills for scaffolding AI agent harnesses: ReAct, plan-and-execute, self-evolving, reflexion, multi-agent, code agent, RAG memory

Python

#agent-framework#agent-harness#agents#agents-md#ai-agents#claude-code#good-first-issue#harness#llm-agents#multi-agent#opencode#rag#react#reflexion#self-improving#skills

failure mode wanted: streaming tool calls arriving as partial deltas about 2 months ago

help wanted failure-mode

ArttuAn/agent-harness-skills

1

IDE skills for scaffolding AI agent harnesses: ReAct, plan-and-execute, self-evolving, reflexion, multi-agent, code agent, RAG memory

Python

#agent-framework#agent-harness#agents#agents-md#ai-agents#claude-code#good-first-issue#harness#llm-agents#multi-agent#opencode#rag#react#reflexion#self-improving#skills

validator: every skill must appear in the README table about 2 months ago

documentation good first issue agent-friendly

ArttuAn/agent-harness-skills

1

IDE skills for scaffolding AI agent harnesses: ReAct, plan-and-execute, self-evolving, reflexion, multi-agent, code agent, RAG memory

Python

#agent-framework#agent-harness#agents#agents-md#ai-agents#claude-code#good-first-issue#harness#llm-agents#multi-agent#opencode#rag#react#reflexion#self-improving#skills

validator: --json output so CI and agents can consume the result about 2 months ago

enhancement good first issue agent-friendly

ArttuAn/agent-harness-skills

1

IDE skills for scaffolding AI agent harnesses: ReAct, plan-and-execute, self-evolving, reflexion, multi-agent, code agent, RAG memory

Python

#agent-framework#agent-harness#agents#agents-md#ai-agents#claude-code#good-first-issue#harness#llm-agents#multi-agent#opencode#rag#react#reflexion#self-improving#skills

validator: require at least one test_ function in every Required tests section about 2 months ago

enhancement good first issue agent-friendly

ArttuAn/agent-harness-skills

1

IDE skills for scaffolding AI agent harnesses: ReAct, plan-and-execute, self-evolving, reflexion, multi-agent, code agent, RAG memory

Python

#agent-framework#agent-harness#agents#agents-md#ai-agents#claude-code#good-first-issue#harness#llm-agents#multi-agent#opencode#rag#react#reflexion#self-improving#skills

validator: fail when a skill cites a references/ file that does not exist about 2 months ago

enhancement good first issue agent-friendly

ArttuAn/agent-harness-skills

1

IDE skills for scaffolding AI agent harnesses: ReAct, plan-and-execute, self-evolving, reflexion, multi-agent, code agent, RAG memory

Python

#agent-framework#agent-harness#agents#agents-md#ai-agents#claude-code#good-first-issue#harness#llm-agents#multi-agent#opencode#rag#react#reflexion#self-improving#skills

validator: require description and argument-hint in every command frontmatter about 2 months ago

enhancement good first issue agent-friendly

ArttuAn/agent-harness-skills

1

IDE skills for scaffolding AI agent harnesses: ReAct, plan-and-execute, self-evolving, reflexion, multi-agent, code agent, RAG memory

Python

#agent-framework#agent-harness#agents#agents-md#ai-agents#claude-code#good-first-issue#harness#llm-agents#multi-agent#opencode#rag#react#reflexion#self-improving#skills

validator: keep the shared command block byte-identical across all seven commands about 2 months ago

enhancement good first issue agent-friendly

ArttuAn/agent-harness-skills

1

IDE skills for scaffolding AI agent harnesses: ReAct, plan-and-execute, self-evolving, reflexion, multi-agent, code agent, RAG memory

Python

#agent-framework#agent-harness#agents#agents-md#ai-agents#claude-code#good-first-issue#harness#llm-agents#multi-agent#opencode#rag#react#reflexion#self-improving#skills

ArttuAn/agent-harness-skills

Open Issues Need Help