Curated APM marketplace of AI context files for software engineering.
Each package bundles primitives that configure AI coding agents — skills,
prompts, agents, and instructions — following the
Agent Skills and AGENTS.md open
standards. Declare what you need in your project's apm.yml and APM resolves,
fetches, and wires primitives into Gemini CLI, OpenCode, Copilot, Claude Code,
Cursor, and every other supported harness.
apm marketplace add pngdeity/apm-user-repository
apm marketplace update
apm install agent-architect@pngdeity-agent-context
apm compile| Package | Type | Description |
|---|---|---|
| agent-architect | hybrid | Meta-tooling for AI context files — agent-architect skill with knowledge sources and file specifications, context-refinement instruction, update-agents-md prompt, and the Confucius skill extraction agent. |
| ai-agent-orchestration | skill | AI agent orchestration design patterns, production best practices, and multi-agent coordination workflows — sequential, concurrent, group chat, handoff, and magentic patterns with decision matrix, production hardening checklist from Google AI Agent Clinic, and practical orchestration workflow from the Cyrus case study. |
| apm-author-defaults | instructions | APM package authoring bundle — create packages, design skills, and manage AI context files |
| apm-eval-defaults | instructions | APM skill evaluation bundle — evaluate SKILL.md quality, run multi-agent eval pipelines, and enforce CI quality gates |
| apm-migration | skill | 6-phase procedural skill for migrating consumer projects to APM — detection, tribal knowledge recovery, scaffolding, install/compile, git operations, and verification |
| apm-package-author | skill | Meta-skills for APM package lifecycle — standalone package creation (anatomy, primitive selection, authoring, local verification) and marketplace registration (root apm.yml integration, version constraints, publishing) |
| apm-vendor-skills | skill | Procedural skill for vendoring external SKILL.md files into an APM marketplace from non-APM repositories — covers discovery, mapping, adaptation, packaging, registration, and validation. |
| architectural-review | skill | ADR and TDD writing and auditing workflow with structural, logic, and consistency review phases. |
| aur-package-management | skill | Arch Linux AUR package maintenance skills — update packages to new upstream versions, recover from patch failures, and bootstrap new packages. |
| automated-modernization | instructions | Blueprint for the Automated Repository Modernization Engine (ARME) — intelligent discovery, staging strategy, and implementation framework. |
| ci-cd-standards | hybrid | CI/CD pipeline design standards — GitHub Actions best practices, Inversion of Control architecture, CNCF Buildpacks migration reference, and vendor-decoupled CI/CD pipeline design skill. |
| code-review-commons | skill | Common guidelines and constraints for performing high-quality code reviews |
| codecompanion | hybrid | CodeCompanion.nvim quick reference and full documentation reference for AI agents working with the Neovim plugin. |
| config-maintenance | hybrid | Configuration file auditing, modernization, and Gemini CLI maintenance guide. |
| configure-trim-safe-efcore | skill | Configure Entity Framework Core for reflection-free compilation, Native AOT compatibility, and aggressive assembly trimming in Blazor WebAssembly apps |
| context-quality-gate | skill | CI/CD quality gate skill for AI agent context files — enforces structural compliance via skills-ref, trigger accuracy thresholds, and output quality minimums before allowing skill publication through automated PR checks |
| custom-skill-creator | skill | Meta-skill for creating and packaging new AI agent skills. Use when creating a skill, documenting a multi-step process, or configuring agent behavior as a skill module. |
| design-process | instructions | Elon Musk's five-step engineering and design process for driving innovation and eliminating inefficiency. |
| design-systems | hybrid | Design system references — Hyperstudio (monochrome terminal with amber accents) and 099 (terminal aesthetic digital workbench). |
| development-practices | hybrid | Development standards — conventional commits, technical integrity, engineering workflow, git standards, and code style conventions. |
| find-docs | skill | Retrieves up-to-date documentation for any library, framework, or tool via Context7 API |
| freecad-development | hybrid | FreeCAD ACP development standards — virtual environment, toolchain, debugging, quality checks, and ACP client implementation. |
| gh-cli-patterns | skill | Patterns for invoking the GitHub CLI (gh) from agents — structured output, pagination, repo targeting, search vs list, and gh api fallback. |
| golang-core-style | hybrid | Go style fundamentals — core style principles, naming conventions, documentation, and package organization distilled from Effective Go, Google Style Guide, and Uber Style Guide. |
| golang-design | hybrid | Go design patterns and quality — interfaces, functional options, defensive programming, error handling, and testing patterns from Effective Go, Google Style Guide, and Uber Style Guide. |
| golang-language | hybrid | Go language constructs — functions, declarations, control flow, data structures, and generics from Effective Go, Google Style Guide, and Uber Style Guide. |
| golang-runtime | hybrid | Go runtime patterns — concurrency, context propagation, logging, and performance optimization from Effective Go, Google Style Guide, and Uber Style Guide. |
| implement-offline-first-sync | skill | Design offline-first local data storage sync logs with optimistic concurrency to handle SQLite-to-server conflicts |
| init-project-guidance | skill | Scaffold GEMINI.md and AGENTS.md for new or existing projects. Detects tech stack and generates tailored AI agent guidance. |
| local-first | skill | Discover and use local information sources (man pages, --help, apropos, info, system logs, kernel params) before reaching for external documentation. Applies when the agent queries installed CLI tools, system utilities, or infrastructure commands. |
| manage-sqlite-wasm-lifecycle | skill | Safely manage SQLite database file lifecycles and DbContext connections during resets, imports, exports, and restores in Blazor WASM sandboxes |
| microsoft-rust-guidelines | hybrid | Microsoft's Pragmatic Rust Guidelines — safety-critical rules enforced implicitly (unsafe code, soundness, panics, lint overrides, Debug/Display), plus an invocable skill covering universal conventions, library design (interop, UX, resilience, building), applications, FFI, performance, documentation, and AI-oriented design |
| normalize-dataset | skill | Pre-process raw documents (PDFs, transcripts, legacy text) into semantic, LLM-friendly Markdown datasets for RAG and knowledge ingestion. |
| pngdeity-defaults | instructions | Default package bundle for pngdeity projects — aggregates development-practices, ci-cd-standards, and code-review-commons |
| project-management | skill | Project management integration skill for AI agents — interface with GitHub Projects, TODO.md, and other project management tools. |
| rust-defaults | instructions | Default Rust development bundle — Microsoft Pragmatic Rust Guidelines with safety-critical rules enforced implicitly and design guidelines on demand |
| skill-eval-agents | hybrid | Seven specialized sub-agent skills and Go CLI tools for autonomous skill evaluation — struct-validator, trigger-evaluator, trigger-aggregator, quality-evaluator, output-grader, revision-synthesizer, and candidate-selector with bundled Go scripts for CLI invocation, session parsing, benchmark computation, and candidate selection |
| skill-eval-pipeline | hybrid | Multi-agent orchestration pipeline for evaluating and optimizing AI agent SKILL.md files — structural validation, trigger testing across opencode and gemini CLI, output quality measurement with graded assertions, automated revision synthesis, and best-candidate selection |
| ssh-gpg-host | skill | Host-specific SSH and GPG configuration constraints for agent commit signing and authentication |
| technical-documentation | skill | Architecture Decision Record (ADR) and Technical Design Document authoring workflows with templates, verification checklists, and anti-pattern guidance. |
| technical-integrity | skill | Maintain intellectual honesty and technical rigor during software engineering tasks. Use when performing code reviews, responding to peer agent feedback, or defending technical decisions. |
| use-modern-go | skill | JetBrains modern Go syntax guidelines — apply Go 1.0 through 1.26 best practices based on detected project version. Covers slices, maps, strings, loops, errors, context, sync, testing, JSON, and HTTP patterns. |