Development plugins

Dependency graphs, docs, specs, a behavior layer, security scan/resolve, status, and a wrap-up workflow that keeps them in sync.

10 skills

Persistent goal mode for Claude Code: namespaced goal commands plus Stop-hook autonomous continuation backed by markdown state.

9 commands · hooks

Damoov skills for mobile SDK integration, backend registration, platform APIs, trips, statistics, and testing.

10 skills

Routes prompts to a cheaper model when one will do: a heuristic classifier picks a tier and effort, and when that tier is strictly cheaper than the session's own model the prompt auto-delegates to a router-{haiku|sonnet|opus} worker subagent. Plan-mode and multi-part prompts fan out into parallel waves with disjoint-file safety, and /router-report scores the cost saved against the parent model, correction rate, and time-to-results.

2 skills · 3 commands · 4 agents · hooks

Adversarial pull-request review: parallel lenses plus a refutation panel that tries to kill every candidate finding.

1 skill · 9 agents

Persistent, self-organizing memory for Claude Code: cross-project pattern detection, concept formation, activation tracking, and Obsidian sync

5 skills · 5 commands

Deploy and run Shiny apps, APIs, Quarto reports, notebooks, and R, Python, and Julia scripts on Ricochet.

10 skills

Java task agents that compose the Java skills — new endpoint, feature, Kafka consumer, metric, client, validation, refactor, code review, repo scaffold, context update, and the parallel test-writer worker. Requires the java-* plugins.

10 agents

The dark-factory working method as plain skills and roles: size a request before any code, plan it in end-to-end slices, build each slice test first, commit only after its tests, and leave the pull request as the record of the work. No hooks and no scripts. Install it or dark-factory, not both.

6 skills · 4 agents

Skills rebuilt for current models, each with a with/without eval and a measured delta.

9 skills

MCP server that saves 98% of your context window.

7 skills · 1 MCP server · hooks

Nine zipsa skills for requirements, diagnosis, planning, implementation, testing and review.

9 skills

Upgrade Claude from text search to code understanding.

8 skills · 1 MCP server

Build modern Deno apps with Fresh, JSR, and Deno Deploy. Enforces best practices like using JSR over deprecated deno.land/x, includes Fresh/Preact/Tailwind patterns, Sandbox SDK guidance, and deployment workflows.

4 skills · 3 commands · 1 agent · hooks

The active Spec Kit spec in the system prompt, every edit booked against a task (Bash writes too), a proof ladder from specified to passing, and an autopilot that runs Spec Kit's whole loop, gated by your spec review and the tests. Band and pane included.

9 skills

Oxy platform skills for coding agents: custom apps and Oxy Functions, semantic models, agents, automations, Airway pipelines, evals, and the oxyc CLI.

9 skills

Frame a SaaS idea with you, then build it end to end with test-driven, simplified, independently reviewed code.

1 skill · 4 commands · 4 agents

Low-overhead architecture, development, review, and explicit local commit skills.

7 skills · hooks

Portable agile development workflow skill for AI coding agents. Includes GitHub issue planning, implementation, validation, PR review, research, and epic planning.

1 skill · 7 commands

Point at a UI bug in your running Next.js app and turn it into a precise, source-located fix — solo with Claude, or routed through BMAD.

1 skill · 6 commands · 1 agent

newman-mode skill and two Bash guards, layered on pstack

3 skills · 4 agents · hooks

Claude Code authoring tools: create plugins, skills, commands, and introspect Claude Code internals

8 skills

Publish diffs, sessions, plans, prompts, and teaching pages as live claude.ai artifact pages — an annotated diff walkthrough with per-hunk reviewer notes, a reviewer-first session recap, plan HTML republished to a stable URL, a filterable prompt library with verbatim copy buttons, and a teach-artifact explainer — plus product-doc, team-recap, and eng-recap audience framings over the session recap, with a blocking secret-scrub gate and a local-HTML fallback.

5 skills · 3 commands

Review, challenge, and delegate tasks to Kimi Code in an isolated workspace.

7 skills · 1 MCP server

Eight skills for Swift systems on the swift-microservices packages: designing (monolith or microservices), building services, HTTP surfaces and gateways, and reusable libraries, writing server Swift, orchestrating Temporal workflows, delivering, and reviewing.

8 skills

Build and audit myOperator design-handoff screens from Merlin. Run /setup once and it connects Merlin and downloads the design system for you.

7 skills · 1 MCP server

Operate the toolkit repo itself: repo orientation, change control and publish gating, skill-authoring standards, debugging playbook, failure archaeology, plugin schema reference, and research frontier for jabworks/agentic-toolkit.

8 skills

Agent-first, human-in-the-loop workflows for infrastructure maintenance.

6 skills · 1 agent · hooks

Skills for working a Loupe project from Claude Code and Codex: documents, the board, the inbox, live review, the workers, the analyses and discovery.

7 skills · hooks

Blender-first, Unreal-shipped content pipeline for NMG games: freeform asset authoring in Blender via MCP, checked by an asset contract and finalized into Desktop and Mobile variants in Unreal Engine.

3 skills · 5 MCP servers