An open format for packaging a complete AI persona as a portable, version-controlled folder. Built on Agent Skills.
# Add to your Claude Code skills
git clone https://github.com/Gabriel-Operator/domain-specific-agent-harness-specGuides for using ai agents skills like domain-specific-agent-harness-spec.
See how domain-specific-agent-harness-spec compares with popular alternatives.
domain-specific-agent-harness-spec is an open-source ai agents skill for AI coding assistants such as Claude Code, Codex CLI, and ChatGPT, built by Gabriel-Operator. An open format for packaging a complete AI persona as a portable, version-controlled folder. Built on Agent Skills. It has 59 GitHub stars.
domain-specific-agent-harness-spec's catalog security scan is still queued. You can run an instant dependency and prompt-injection check now with the "Scan for vulnerabilities" button above.
Clone the repository with "git clone https://github.com/Gabriel-Operator/domain-specific-agent-harness-spec" and add it to your Claude Code skills directory (see the Installation section above).
domain-specific-agent-harness-spec is primarily written in MDX. It is open-source under Gabriel-Operator on GitHub, so you can review or fork the full source.
Yes. SkillsLLM lists many other AI Agents skills you can browse and compare side by side. Open the AI Agents category from the badge at the top of this page, or use the Related Skills and comparison links further down to weigh domain-specific-agent-harness-spec against similar tools.
No comments yet. Be the first to share your thoughts!
Available after the catalog security scan passes.
⚠️ Third-Party Software Notice
This skill is third-party open-source software developed and hosted independently on GitHub. SkillsLLM lists this project for discovery and does not control or maintain the underlying repository.
Any security checks, ratings, or warnings displayed by SkillsLLM are automated and limited in scope. They do not constitute a security certification or guarantee that the software is safe, error-free, or free from malicious code, vulnerabilities, compromised dependencies, or prompt-injection risks.
Review the source code, permissions, dependencies, and configuration before installing or running any third-party skill. Use is at your own risk. To the maximum extent permitted by applicable law, SkillsLLM is not liable for losses arising from third-party software.
The deep catalog scan for this skill is still queued. Run an instant dependency check now instead.
domain-specific-agent-harness-spec vs hermes agent
See comparison
A persona is a versioned Git definition of a domain, its teams, and the playbooks they use. It also carries shared data definitions, rules, and evals. The runtime verifies those definitions and binds workspace resources. The agent harness runs the agent loop and coordinates model calls and tool use.
A portable playbook declares a versioned workflow, its skills, and the capabilities it needs. A compatible harness verifies the package, resolves the destination’s bindings and permissions, and executes it when invoked by a team goal. Portable playbook contract · JSON and YAML examples
Follow the illustrated walkthrough.
| Term | What it means |
|---|---|
| Persona | The Git definition of a domain model, its teams, and the playbooks they use. |
| Playbook | A reusable workflow with declared inputs, outputs, operations, and approval rules. |
| Goal | A desired outcome within a team that can invoke one or more playbooks. |
| Runtime | Loads and verifies a persona, connects workspace resources, and coordinates execution. |
| Harness | Runs the agent loop and coordinates model calls and tool use. |
| Agent | A worker that uses instructions and tools. A team can include several agents. |
| Workspace | One installation's accounts, records, permissions, and run history. |
The root SKILL.md follows Agent Skills. A persona adds declarative definitions and a registry of pinned resources. This draft covers persona import and runtime behavior, including reporting which steps can run. How a runtime connects to its harness depends on the implementation.
Existing paths and identifiers such as assets/persona-evals.json and teams[].bots still apply.
persona-lane/
├── SKILL.md
├── assets/
│ ├── chat-config.json
│ ├── chat-app.json
│ ├── ontology.json
│ ├── persona-evals.json
│ ├── studio-teams.json
│ └── data-feeds/
├── references/
│ ├── registry.json
│ ├── workspace.json
│ ├── lists/
│ ├── pipelines/
│ └── workflows/
└── skills/
The folder reference explains which files each profile uses and which resources belong to the workspace. The implementation guide maps the architecture to the files and includes example JSON.
The draft marks features Extracted (used in Gabriel today), Proposed (new in this draft), or Planned (still to be specified). The core format and registry are extracted. Runtime binding and presence are proposed. A complete set of interoperable schemas and a shared team orchestration contract are planned.
| Page | Start here for |
|---|---|
| Specification | Folder layout, stable keys, registry, definition files, and runtime rules |
| Implementation guide | Architecture-to-file map, example JSON, and runtime binding |
| Apps and connectors | Connector operations, shared records, app packages, and live sync |
| Concept reference | Personas, teams, goals, connectors, apps, and runtime behavior |
| Local AI | Bind a harness to a model running on a Mac mini, DGX Spark, or another private inference host |
| Playbooks (skills) | Executable definitions, reusable skill files, and requirements for running them in a team |
| Skill catalogue | Gabriel Operator authoring skills and individual action, node, and component references |
| Examples | The Lane Studio tour and 17 persona architectures |
| Terms and standards | Vocabulary and how Agent Skills, MCP, A2A, and Standard Agents fit |
The format comes from Gabriel Operator. The Portable Persona Runtime installer deploys published personas to a self-hosted runtime appliance. Check the runtime's capability report for the steps it can execute, especially offline.
For a coding agent, start with:
Read https://gabriel-operator.github.io/domain-specific-agent-harness-spec/llms.txt before authoring a persona or implementing a runtime.
Requires Node.js 22.23 or later.
npm ci
npm run dev -- --background
npm run build
The site uses Astro and Starlight. Pages live in src/content/docs/, diagrams in src/components/, and styles in src/styles/. The build writes dist/. The GitHub Pages workflow sets SITE_BASE=/domain-specific-agent-harness-spec. public/llms.txt indexes the source for coding agents.
See Contributing to report an issue or propose a change. Released under the Apache License 2.0.