Licence-vetted systems-engineering knowledge-pack skills for coding agents, reconstructed from genuinely open sources. Tooling MIT; pack content under each source's own licence.
Agentic software-delivery loop: turn specs into a dependency-ordered GitHub Issues backlog, then let agents claim, implement, adversarially review, patch, and merge tasks in isolated git worktrees behind a single correctness gate.
★not rated 5 2mo agoA
tokens not measured
originalMIT
Azure Kubernetes Service (AKS) agent skills and MCP integration: troubleshoot live clusters, optimize cost, assess AKS Automatic readiness, operate GPU and model-inference workloads, capture packet-level evidence, and design clusters. Deep Day-2 AKS operations that complement the broader Azure Skills plugin.
★not rated 5▲
+1 6d agoA
tokens not measured
originalMIT
Analyse de profil GitHub orientée objectif : eval README, détection marqueurs IA, signaux recruteur, conseils keywords et SEO. Né du live TPC GitHub Roast.
Linear integration: kanban board, cycle management, issue archive/delete, and comment edit/delete. Uses official Linear MCP for standard CRUD operations.
★not rated 5 6mo agoA
tokens not measured
originalMIT
A bundle for product docs and the tasks that come out of them. Write a spec against a format and verify it before it ships; turn a request thread into a task's context table; file that task in your engineering tracker and keep the two sides reconciled. Where things are stored is decided by config.
Furl context compression for Claude Code: bundles the Furl MCP server (furlcompress / furlretrieve / furlstats / furlpurge / furlsearch / furllist) and a fail-open PostToolUse hook that shrinks large tool outputs before they land in context. Originals stay byte-exactly retrievable via > markers.
★not rated 5▲
+1 2d agoA
tokens not measured
originalApache-2.0
Session-state handoff for Claude Code: hooks auto-snapshot at SessionEnd/PreCompact and auto-load at SessionStart. Rebuilds lost handoffs from per-turn backups, keeps rotating history, nudges at 40% context, and HMAC-signs handoffs so cloned repos can't inject trusted rules.
★not rated 5
changed 4d agoA
tokens not measured
originalMIT
A plan/execute/verify loop for a single developer in Claude Code, with adversarial review at every gate: plans and diffs get refuted by a fresh Claude subagent or a panel of outside models (OpenAI, Gemini, DeepSeek), not just approved. State lives in files, not the conversation, so you clear aggressively and keep both.
★not rated 5
changed 10d agoA
tokens not measured
originalMIT
Local web dashboard and CLI for /.claude config visualization — measured per-session token cost, install-vs-used gaps, and a review-first cleanup script for plugins, agents, skills, hooks, and MCP servers.
★not rated 5 1mo agoA
tokens not measured
originalMIT
Public-metadata domain intelligence for Microsoft 365 / Google Workspace tenant identification, email security observation (DMARC, DKIM, SPF, MTA-STS, BIMI), SaaS fingerprinting, certificate-transparency findings, and posture analysis. Bundles the recon MCP server and its usage guidance.
★not rated 5
changed 2d agoA
tokens not measured
originalApache-2.0
An opinionated governance layer for Claude Code, built for solo developers: autonomous sprint execution in isolated git worktrees, mandatory TDD, quality gates, pre-PR code review, and structured planning — workflow enforced by hooks, not just rules.
★not rated 5 1mo agoA
tokens not measured
originalMIT
Your AI on rails. On your terms. An open-source protocol engine that puts every process on rails — enforced as it happens, so your AI can't fake the work.
★not rated 5 1mo agoA
tokens not measured
originalMIT
Skills that drive the VCF Automation Orchestrator MCP server safely: discovery-first authoring, package-first publishing, and guided troubleshooting via the server's vcfa- prompts and tools.
★not rated 5
changed todayA
tokens not measured
originalApache-2.0
Session intelligence for Claude Code: search, analyze, diff, and resume past sessions with a web dashboard.
★not rated 5 4mo 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: