Find plugins

by Scott Solo

Ask ACQ AI (Mozi) from Cowork or Claude Code with your own docs as the context: Claude writes the question, sends it through your own signed-in browser on your yes, refines the answer with follow-up questions, and brings the mechanics home with the edits they imply.

by Actian UX Team

Actian Design System toolkit: generate wireframe flows with prototype wiring, create component briefs, propose component-scale designs as reasoned documents, audit designs, compare flows. 7 skills (with tiered generation: recognized / adapted / improvised), 10 agents, 155 tokens (3 themes, 8 collections), WCAG 2.2 AA. DS knowledge vendored from volivarii/actian-ds-knowledge.

action-full-pass is an open-source Agent Plugins plugin indexed from GitHub with 1 skill and 2 MCP servers. Review its manifest and source before installing.

action-mixed is an open-source Agent Plugins plugin indexed from GitHub with 1 MCP server. Review its manifest and source before installing.

action-single-rule is an open-source Agent Plugins plugin indexed from GitHub. Review its manifest and source before installing.

action-subdirectory is an open-source Agent Plugins plugin indexed from GitHub. Review its manifest and source before installing.

action-unable is an open-source Agent Plugins plugin indexed from GitHub with 1 MCP server. Review its manifest and source before installing.

by jabworks

Codex skill with stack and style priors for adapting generated work to the jabworks conventions.

by David Christensen

Development-workflow skills: design, TDD, adversarial review, shipping, and campaign orchestration.

by Shmulik Davar — BrAIght Wave

Agent engineering for teams building AI agents and LLM features: architecture patterns, tool design, eval suites, prompts as versioned code, and runtime guardrails.

by Shmulik Davar — BrAIght Wave

Agentic construction at the system level: choosing and composing execution frameworks with one owner per slot, long-running and multi-agent work design, small-batch delivery policy, and legacy modernization strategy with behavior parity.

by Shmulik Davar — BrAIght Wave

Context engineering for coding agents: CLAUDE.md / AGENTS.md authoring and audits, codebase maps, context-window budgeting, codifying conventions as skills, hooks or tests, MCP integration planning, and evidence-based skill library management.

by Shmulik Davar — BrAIght Wave

ADLC foundations: readiness assessment on the DORA AI capabilities, SDLC-to-ADLC workflow mapping, agent autonomy levels, AI stance policy, ROI metrics, AI champions program, role transitions, and comprehension debt.

by Shmulik Davar — BrAIght Wave

Governance and security for agentic development: OWASP Agentic Top 10 threat modeling, auto-mode-era agent permissions, guardrail hooks, compliance mapping (ISO 42001, NIST AI RMF, EU AI Act, SOC 2), extension vetting, and governance for non-engineers building with AI.

by Shmulik Davar — BrAIght Wave

Intent specification for coding agents: Agentic PRDs, spec-driven development, testable acceptance criteria (Given/When/Then, EARS), spec clarification, task decomposition, and architecture guardrails (ADRs, invariants, dependency rules).

by Shmulik Davar — BrAIght Wave

Operating the ADLC: release gates, Continuous AI background workflows, AgentOps observability, AI cost management, a learning loop from rejected agent work, and blameless agent incident reviews.

by Shmulik Davar — BrAIght Wave

Verification of agent-built software: review capacity design for the agent-PR bottleneck, behavioral testing, tests derived from specs, alignment review of agent PRs, hallucination checks, and a Definition of Done for agent work.

by Long

Lightweight, evidence-first software delivery for Codex. 中文优先的 AI 软件交付工作流。

by Jordan Richlen

Compile deterministic, content-hashed agents from small behavior modules: a skill turns fuzzy intent into a typed AgentQuery, a stdlib-only kernel resolves modules, expands dependencies, fails closed on conflicts and over-ceiling effects, and emits an immutable AgentImage plus a rendered harness-native subagent — never inventing behavior text without provenance.

by Mustafa Qasim

Check whether non-trivial engineering work needs Agent Council model routing and evidence gates.

by Javi

Set up and update portable agent workspaces, coordinate scoped unattended work, and review UI changes with visual evidence.

by Anton Novoselov

Multi-stage agent pipeline for feature development with deterministic quality gates: a specifier writes Gherkin acceptance criteria, a coder implements to green tests, a cleaner drives every function under a per-function CRAP threshold (complexity^2 * (1 - coverage)^3 + complexity), a hardener kills surviving mutants via mutation testing

by Andrew Maston

Investigate agent transcripts — locate, summary, errors, retries, pivots, report.

by Agent Issues contributors

Agent Issues workflow skills and MCP tools.