From truefoundry
Guides users through the TrueFoundry Agent Registry UI for creating, editing, publishing, and testing agents. Use when managing agents via the dashboard.
How this skill is triggered — by the user, by Claude, or both
Slash command
/truefoundry:agentsThis skill is limited to the following tools:
The summary Claude sees in its skill listing — used to decide when to auto-load this skill
<objective>
Guide the user through the TrueFoundry Agent Registry UI.
Agent authoring and interaction are dashboard workflows. Do not try to create, publish, edit, invoke, or delete agents through the API.
Use this skill when the user wants to:
tfy login is complete. If not, use truefoundry-onboard.Do not ask for API keys for agent authoring.
Direct the user to:
TrueFoundry Dashboard -> Agent Registry -> Create New Agent -> Build Agent on TrueFoundry
In Agent Playground:
Save Agent.Save New Agent for a new agent or Save New Version to publish a new version of an existing agent.Direct the user to:
TrueFoundry Dashboard -> Agent Registry -> Manage
Editable fields include:
After edits, test in the playground and save a new version.
In Agent Playground:
+ next to MCP Servers.If the MCP server does not exist yet, use truefoundry-mcp-servers first.
In Agent Playground:
+ next to Skills.SKILL.md.Use preload only for short, always-relevant skills. Leave it off for long or situational skills.
If the skill does not exist yet, use truefoundry-skills-registry first.
Use the agent edit form to grant:
For broader teams and roles, use truefoundry-platform.
For agents running outside TrueFoundry:
TrueFoundry Dashboard -> Agent Registry -> Create New Agent -> Remote Agent
Collect:
Then guide the user through the UI form.
Do not delete agents for the user. If deletion is requested, direct them to the dashboard:
TrueFoundry Dashboard -> Agent Registry -> Manage -> delete from the UI
<success_criteria>
</success_criteria>
npx claudepluginhub truefoundry/skills --plugin truefoundryDeploys, evaluates, and manages Microsoft Foundry AI agents: Docker builds to ACR, hosted agent creation, batch evals, prompt optimization, tracing, troubleshooting.
Guides developers to create new AgentCore agent projects on AWS: framework selection (Strands, LangGraph), project scaffolding, first deploy, and invocation. For beginners or 'agentcore create'.
Guides creation and configuration of autonomous agents for Claude Code plugins, covering frontmatter, triggering descriptions, system prompts, tools, teams, permissions, and best practices.