#llm

AgentS Multi-Agent Harness

agent-smith

AgentS multi-agent harness for VSCode: OpenAI-compatible endpoints, Claude Code and Antigravity agents, streaming runs, approvals, and diffs.

Claude Code Warden

NovisLabsIsrael

Token-usage insight and safety prompts for Claude Code: a live status-bar meter, context-window gauge, per-session budgets, and Allow/Deny prompts for risky commands.

MoA Bridge - Mixture of Agents for Copilot

dudali095

Multi-LLM orchestration for VSCode Copilot Chat. Runs a 5-role pipeline (Planner → Recon → Refs → Aggregator → Actor) via the native vscode.lm API. Supports @moa chat participant, LM tools, and preset model groups.

Make agents control your editor

AskFirst

EKhurgin

Refines your AI prompts by asking clarifying questions first, then rewriting the prompt. Runs locally via Ollama.

Derba

krave

Explicit, checkbox-driven context for AI chat — you choose what the model sees

RouterBase

RouterBase

Configure RouterBase as an OpenAI-compatible API gateway from VS Code.

Silvanexum

Silvanexum

Spend control for AI agents, in your editor — budgets & enforcement, spend attribution, the agent→agent mesh hop ledger, fleet drift, and signed-trace replay.

Clipboard To Markdon

gk62

Save clipboard text as a Markdown file with an automatically generated filename.

PII Guardian

DevSpark

Filter PII from code and chat before sending to LLMs

SANIX — Agentic Neural Intelligence eXecutor

Istiak-Ahamed

SANIX in your editor — 22 specialized agents, streaming chat, inline edit (Cmd+K), diff applier, V16 speed-first intelligence pipeline.

nieeek.I. Assistant

nieeeko

Agentischer KI-Coding-Assistent mit vollem Repo-/Dateizugriff, Terminal, Multi-Provider (nieeeki/Ollama, OpenAI, Anthropic, OpenRouter, Groq).