agent-engineering-toolkit
v1.4.0by haydarrjr · author: Önder Türkan
Open-source engineering skills for repository work, Python, host-native MARGOS orchestration, ReThinking audits, and Agent Plugins authoring.
Package we checked
plugin.json sits at the package root in haydarrjr/agent-engineering-toolkit, which is the plugin directory the 1.0.0 spec describes. Its $schema is https://agent-plugins.org/schemas/1.0.0/plugin.schema.json. The manifest lists 5 skills (agent-plugins-author, margos, python-engineering-harness, rethinking, software-craft) and no MCP servers.
What agent-engineering-toolkit does, in the publisher's words
Agent Engineering Toolkit (AET) is an open-source, skills-first engineering plugin for agent-assisted software development. It combines focused repository engineering, a Python-specific engineering harness, explicit coordination/governance tools, repository instruction audits based on OpenAI's current ReThinking guidance, and Agent Plugins authoring/validation in one portable package.
AET is intentionally small at activation time: the host sees five narrow skills and loads detailed references or scripts only when the task needs them.
> AET is an independent open-source project. It is not an OpenAI or GitHub product and is not endorsed by either company.
From the project README, punctuation lightly normalized · Open the README on GitHub
- Skills
- 5
- MCP servers
- 0
- Stars
- 2
- License
- Apache-2.0
- Repo created
- 2026-09-16
- Last pushed
- 2026-09-21
- Publisher type
- User
- Version
- 1.4.0
Links
Skills · 5
- agent-plugins-author
- Use only when the user explicitly asks to create, audit, port, update, package, validate, or reconcile Agent Plugins, plugin manifests, skill bundles, marketplaces, or Codex/Copilot plugin compatibility.
- margos
- Use only when the user explicitly requests MARGOS coordination, optional delegation, host-capability decisions, or bounded runtime/profile changes.
- python-engineering-harness
- Use when Python repository code, tests, packaging, or Python-specific architecture is material to the requested engineering work.
- rethinking
- Use only when the user explicitly requests an OpenAI ReThinking/GPT-6 Astra audit or repair of repository instructions, skills, prompts, or agent configuration.
- software-craft
- Use when repository-grounded software work is cross-language or has no narrower language-specific owner.
Descriptions come from the frontmatter of each SKILL.md, punctuation lightly normalized.
Category
Other. Assigned by this directory. Agent Plugins 1.0.0 has no category field, so no manifest declares one.
Keywords
software-engineering · python · codex · agent-skills · agent-plugins · code-review · debugging · orchestration · copilot · open-source-maintenance
Related by publisher and keywords
Author, audit, and package Agent Plugins. Use for portable manifests, skills, marketplaces, or Codex/Copilot IDE compatibility.
Agent skills for research, App Store readiness, Codex model routing, repository and Swift review, PR feedback triage, and safe branch cleanup.
Portable Rootloom Agent Skills for inspectable code changes and evidence-backed review with bounded project and artifact context.
Hosted CentricMem librarian: organise and retrieve Markdown memory via host MCP.
Multi-agent engineering orchestration with planning, parallel execution, evaluation loops, debugging, TDD, and executive/technical review skills.
Risk-first PR/branch review: blast radius, agent-authored checks, call-graph pincer, structure.
[](https://agentpluginsdirectory.com/plugins/agent-engineering-toolkit)✓ Verified . We fetched the manifest from GitHub and checked it against the official Agent Plugins 1.0.0 schema at agent-plugins.org.