Automate Claude Code SKILL.md improvement with agent-driven eval loops, binary assertions, and git commits for steady skill gains
npx add-skill Nikkiunitary185/autoimprove-ccAutomate Claude Code SKILL.md improvement with agent-driven eval loops, binary assertions, and git commits for steady skill gains
Source: https://github.com/Nikkiunitary185/autoimprove-cc
Discovered during the daily awesomeskills.dev agent-skill hunt.
LLM 驱动的 DevOps Copilot 平台 —— 多租户 Agent 分诊告警、review PR、修 bug、响应 oncall,插件化、完整审计、默认 dry-run + 人工审批。 LLM-powered DevOps copilot platform. Multi-tenant agents triage alerts, review PRs, fix bugs, and respond to pages — plugin-based, audit-logged, dry-run + human approval by default.
A repo of AI agent skills
bstack P9 — CI watcher + productive-wait primitive. Replaces sleep-based CI waits with an event-driven control loop, drains a context-scoped wait-queue while CI runs, classifies failures, and self-heals known categories. Merge authorization stays with the control metalayer.
Agent Skills to help developers using AI agents with Supabase
The API layer for AI agents. Dashboard + 22K APIs + 18 Direct Call providers. MCP native.
A Claude Code skill that performs automated, read-only EKS cluster upgrade readiness assessments. Know what will break before you upgrade.