steve

executus

Batteries-included base for building LLM agent harnesses in Go (above majordomo). Vibe-coded.

Updated 2026-06-30 15:56:48 +00:00

gadfly

🪰 Agentic adversarial code reviewer for Gitea Actions — hunts real bugs across a swarm of specialist lenses, verifies each against the checked-out repo, and posts one advisory PR comment. Provider-agnostic via majordomo. Advisory only, never blocks a merge.

Updated 2026-06-30 15:41:04 +00:00

majordomo

Go 0 0

Provider-agnostic Go LLM client — one interface over Ollama (local/cloud/foreman), OpenAI, Anthropic, and Google, with failover chains, env-DSN aliases, tool calling, and structured output. The engine behind gadfly.

Updated 2026-06-29 03:09:32 +00:00

llama-swap

Go 0 0

Reliable model swapping for any local OpenAI/Anthropic compatible server - llama.cpp, vllm, etc

golang llama llamacpp localllama localllm openai openai-api vllm

Updated 2026-06-28 16:56:21 +00:00

gadfly-reports

Go 0 0

Updated 2026-06-28 00:39:45 +00:00

gadfly-mcp

Go 0 0

Updated 2026-06-27 03:55:25 +00:00

foreman

Go 0 0

🪓 Small always-on Go daemon that fronts one Ollama target — turns it into a queued, observable job endpoint (model-swap serialization, job IDs, progress webhooks). Speaks native Ollama on the wire, so it's a drop-in target for any Ollama client.

Updated 2026-06-27 00:33:42 +00:00

go-llm

Go 0 0

Abstraction layer interface for various similar LLM services

Updated 2026-05-31 22:44:11 +00:00

go-extractor

Go 0 0

Updated 2026-05-15 22:39:41 +00:00

answer

Go 0 0

Updated 2025-11-11 05:07:52 +00:00

aislegopher