AI Engineering Army — AI skill for Claude Code
Portable Agent Skills framework for Claude Code, Codex, and OpenCode.
How to install AI Engineering Army
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open himanshu231204/ai-engineering-army and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What AI Engineering Army does
Portable Agent Skills framework for Claude Code, Codex, and OpenCode. Nine MIT-licensed skills for delegated reviews, evidence-based verification, and bounded repair. Not an agent, orchestrator, or hosted service.
Alternatives in AI
- Hindsight — State-of-the-art long-term memory for AI agents by Vectorize 6.7k ★
- Open Dots — Open-source, self-hosted AI agent workspace and alternative to OpenAI Dots, Meta Muse, Grok Bot, Instinct, Man 5.5k ★
- Pi Agent VS OpenCode — Customization & Control Comparison — Pi v0.52+ vs OpenCode v1.1+ (Feb 2026) Thesis: Pi and OpenCode are both MIT-licensed, open-source, model-agnos 526 ★
README
AI Engineering Army (AIEA)
AIEA is an open-source, portable Agent Skills framework for Claude Code, Codex, and OpenCode. It improves engineering workflows in the coding host people already use. It is not a standalone coding agent, orchestration engine, or application.
**Source version:** `1.3.3` (public release) · **License:** MIT
Start with the [developer guide](docs/getting-started.md) for installation, your first task, and troubleshooting. Use explicit bootstrap activation for the preview. Packaging and bounded Codex Work exercises have been checked; automatic activation and completed native Claude/OpenCode model workflows remain unverified. See [the evidence and limits](docs/verification-status.md).
**What this plugin runs.** It ships Markdown instructions and two static adapter files. It contains no hooks, no MCP servers, no commands, no agents, and no executable scripts, and it makes no network requests or telemetry. The only file that executes is the optional OpenCode 2 registration module, which reads its own bundled skill files and contributes skill definitions. `AGENTS.md` and `AGENTS.md` at the repository root are contributor instructions for people working *in* this repository; they are not loaded as plugin context. The plugin ships its own user-facing context in the `aiea-bootstrap` skill.
What v1 includes
- A Claude Code marketplace/plugin adapter and standalone plugin ZIP.
- A Codex plugin/local-marketplace ZIP and an optional OpenCode 2 skill-registration plugin.
- Seventeen portable skills: ten core workflow skills and seven read-only specialist contracts.
- A dependency-free Python CLI to validate and list skills and build ZIP packages for supported hosts.
- Host compatibility guidance for native discovery and packaged skill locations.
- Security, contribution, release, evaluation, and final architecture documentation.
- A small evaluation fixture for the separate ArmyBench project.
Claude Code plugin
AIEA includes a Claude Code plu
Related Skills
Edu
Edu — installable, LLM-agnostic agentic harness: a self-improving second brain, an orchestrator with identity,
Claude Code Blueprint
Battle-tested reference architecture for Claude Code power users. Specialized agents with model tiering, nat
Synthcode
SynthCode — the open-source Claude Code alternative. Battle-tested agent framework, model-agnostic, TypeScript
Resume Tailor Skill
A Claude skill for tailoring resumes, cover letters, and interview prep to any job posting, in any field. Made
Modelrudder
Experimental local LLM router for OpenAI Codex CLI and Claude Code. Automatically selects models per task usin
Learn Agent Engineering
12-course AI agent curriculum: Claude Code to production systems. Six languages. MIT licensed.
Related Agents
ZeroClaw Android
Run AI agents 24/7 on your Android phone. Native Rust core, 25+ providers (OpenAI, Claude, Gemini, Groq, DeepS
Compliance Fixer
Agent-agnostic compliance remediation prompt. Implements bounded SOC 2 or HIPAA fixes with verification, rollb
Fidelity QA
팩 렌더를 원본 레퍼런스와 페이지 단위로 대조해 원문 충실도 7축을 채점하고 합격·반려를 판정하는 검증 게이트. Codex/GPT 이종 모델 교차검증을 병행한다. 팩을 직접 수정하지 않고 repai