domain-specific-agent-harness-spec
llms.txt16 starsChanged 4 days agoApache-2.0View on GitHub
Draft 0.1. Inventory the organization’s apps with a domain specialist and map their records to an ontology. Custom apps read and chart the mapped data. Agents read and update authorized apps or shared records while the team runs a goal on its harness. Specialist approvals and corrections provide evidence for reviewed improvements to definitions and evals.
In this repositoryOne other file it gives its agents.
- AGENTS.mdAGENTS.md
The file
Files
# Domain-Specific Agent Harness Spec > Draft 0.1. Inventory the organization’s apps with a domain specialist and map their records to an ontology. Custom apps read and chart the mapped data. Agents read and update authorized apps or shared records while the team runs a goal on its harness. Specialist approvals and corrections provide evidence for reviewed improvements to definitions and evals. - [Overview](https://raw.githubusercontent.com/Gabriel-Operator/domain-specific-agent-harness-spec/main/README.md): How personas, runtimes, and harnesses fit together. - [Homepage story](https://gabriel-operator.github.io/domain-specific-agent-harness-spec/#why-a-domain-specific-harness): Why a team needs shared rules and how a domain-specific harness carries a goal from plan to published records. - [Specification](https://raw.githubusercontent.com/Gabriel-Operator/domain-specific-agent-harness-spec/main/src/content/docs/specification.mdx): Folder structure, SKILL.md, definitions-only rule, resource keys, registry, and runtime rules. - [Implementation guide](https://gabriel-operator.github.io/domain-specific-agent-harness-spec/implementation/): Map the architecture to definition files, workspace behavior, and validated example JSON. - [Apps and connectors](https://raw.githubusercontent.com/Gabriel-Operator/domain-specific-agent-harness-spec/main/src/content/docs/apps.mdx): Connector requirements, steps, lists, pipelines, app packages, live sync, and presence. - [Concept reference](https://gabriel-operator.github.io/domain-specific-agent-harness-spec/concepts/): Individual concept pages covering core definitions, execution and governance, data and integrations, and infrastructure and distribution. - [Governed execution](https://gabriel-operator.github.io/domain-specific-agent-harness-spec/concepts/governed-execution/): Human decisions, isolated rehearsal, exact effect approval, and live receipts. - [Skill catalogue](https://raw.githubusercontent.com/Gabriel-Operator/domain-specific-agent-harness-spec/main/src/content/docs/skills.mdx): Individual Gabriel Operator skills and child references. - [Examples](https://raw.githubusercontent.com/Gabriel-Operator/domain-specific-agent-harness-spec/main/src/content/docs/examples.mdx): Studio tour, a prospect workflow, and 17 persona architectures. - [Terms and standards](https://raw.githubusercontent.com/Gabriel-Operator/domain-specific-agent-harness-spec/main/src/content/docs/standards.mdx): Terminology, Agent Skills, MCP, A2A, and the Standard Agent Specification. - [Agent Skills specification](https://agentskills.io/specification): Format for the root SKILL.md and bundled skills.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.