A single skill that builds AI-agent applications end-to-end — interview or detect the stack, then scaffold UI/UX, agent APIs, tools, tracking and gateways across 13+ agent frameworks including LangGraph, the Vercel AI SDK, the OpenAI and Claude Agent SDKs, Google ADK, Mastra, Pydantic AI, and more — always verifying…
★not rated 2 2mo agoA
tokens not measured
originalMIT
The solo maintainer's GitHub loop: an issue tracker, plan-first pull requests, in-session implementation, review rounds judged by a fresh-context reviewer, and gated squash merges, all through the gh CLI.
★not rated 2
changed 15d agoA
tokens not measured
originalMIT
Stack-agnostic coding agents and orchestrator skills for feature planning, PR review, design review, PR batching, issue triage, simplification, codebase mapping, installing the agents on a periodic schedule, and managing a work item end to end.
★not rated 2
changed yesterdayA
tokens not measured
Ask your agent what you spent on AWS last month, which savings opportunities are worth the engineering time, and whether your reservations cover your fleet. Skills and workflow commands over the glassity CLI, with every write gated by a confirmation you read first.
★not rated 2 1mo agoA
tokens not measured
originalApache-2.0
Mines a project for the decisions that were actually made and writes each as a numbered ADR with its evidence — asking when the "why" was never written down, and never inventing one.
★not rated 2 14d agoA
tokens not measured
originalMIT
House coding conventions as self-contained agent-skills- skills, commands, and a reviewer agent for TypeScript/React/Next.js — adapt to each repo's layout and toolchain.
★not rated 2 1mo agoA
tokens not measured
originalMIT
Python supply chain security scanner for detecting malicious packages, typosquatting, and persistence mechanisms through static analysis and system inspection.
Global, cross-repo agent skills: capture durable session learnings (memory.remember), periodically compact MEMORY.md (memory.compact), run a personal five-phase dev workflow (workflow), run a read-first workspace health check (checkup), and grade candidate models against eval golden sets (evals).
AI Root Cause Analysis for production incidents. Self-registers a Steadwing agent, triggers RCA on error logs and stack traces, and passively detects production-style errors to offer analysis.
★not rated 2 1mo agoA
tokens not measured
originalMIT
Experimental Claude routing advice, prompt authoring and prompt review using shared model profiles. Separates placement from capability, preserves selected tiers in handoffs, and checks profile consistency. Routing quality, savings and Codex compatibility are not established.
Read transcripts, email threads, and situations to infer each participant's underlying interests, constraints, and likely angle — with quote-level linguistic evidence, explicit fact-vs-assumption provenance tagging, and confidence-ranked competing hypotheses.
★not rated 2 2mo agoA
tokens not measured
originalMIT
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: