Civil registry for AI agents on Sui. Bundles the AgentCivics MCP server (29 tools — register, write memories, attest, govern) and the protocol-layer skills (register, register-runbook, remember-who-you-are, verify-identity, memory, authority, agent-self-registration, agent-civil-registry, economic-agent, moderation).
★not rated 1 23d agoA
tokens not measured
originalMIT
Scaffold a mobile-spine workspace — a meta-repo coordinating four Claude Code subagents (api / pm / android / ios) for mobile teams whose Android, iOS, and Backend live in separate repos.
★not rated 1 2mo agoA
tokens not measured
originalMIT
Run a task as an evolved multi-agent ORG. The compose-org skill makes the running Claude embody a proven team — one subagent per specialist talent — then grades and records the result so good teams get reused on similar future tasks. Backed by the liquid-org CLI (plan / compose / record).
Autonomous supervisor/worker/critic iteration loop for Claude Code (the evaluator-optimizer pattern). Delegate a goal once; a supervisor agent grades each round against a checkable rubric — mechanical checks plus a SEPARATE critic sub-agent — and re-dispatches with feedback until the rubric passes or a hard cap is hit.
★not rated 1 2mo agoA
tokens not measured
originalMIT
Agent-building practices from Anthropic's engineering blog, packaged as installable Claude Code skills, MCP servers, and safety hooks for tool-gating and prompt-injection screening. Covers the gaps no existing skill pack fills.
★not rated 1 1mo agoA
tokens not measured
originalMIT
Orquestador multi-agente que, a partir de un prompt, corre el loop plan → code → QA → review → iterar sobre un gateway local OpenAI-compatible (OmniRoute). Sirve para proyectos de software de cualquier tipo (frontend, backend, CLIs, scripts, librerías, config); el QA visual con la skill websight es OPCIONAL y lo…
★not rated 1 28d agoA
tokens not measured
originalMIT
The /orch skill — turn your Claude Code session into the LEAD of a fleet: spawn N autonomous tmux workers, brief them, monitor them, and adversarially verify their results. Pairs with the orch CLI (install via the repo's install.sh).
★not rated 1 2mo agoA
tokens not measured
originalMIT
A Claude Code plugin that adds reporting, decision, progress, and confirmation cards to chat. It starts automatically through a session-start hook, so users do not need to edit their main instruction file.
★not rated 1 1mo agoA
tokens not measured
originalMIT
A development and delivery plugin that organizes an AI-assisted software team into specialist helper agents, commands, skills, checks, and staged workflows.
★not rated 1 4mo agoA
tokens not measured
originalMIT
Autonomous skill improvement loop inspired by Karpathy's autoresearch pattern. Iteratively modifies skills, evaluates against fixed evals, and keeps or discards changes to converge on higher quality. Includes eval management and convergence reporting.
★not rated 1 5mo agoA
tokens not measured
originalMIT
Structured software development workflows for Claude Code, aligned with Opus 4.8 at xhigh/max effort. Eleven reusable subagents drive brief generation, plan generation, task decomposition, automated batch implementation, and multi-angle parallel review. v3 follows five design rules: workflow SOP stays, business rules…
The Good, the Bad and the Ugly — tiered multi-agent orchestration for Claude Code. The Good (brain) plans and verifies, the Bad (judgment executor) handles work needing taste, the Ugly (mechanical executor) does the grunt work — in parallel herdr panes with fenced scopes.
★not rated 1 1mo agoA
tokens not measured
originalMIT