workday-agent-dialogue
Use after Workday routing has already selected an agent, before you send the agent message. Batch related parts into one message and confirm an agent write with the user. Do not use for a catalog tool call. Do not choose an agent or an agent_id. Load the workday skill first.
Install
With your agent
Install OMGSkills skill "workday/everywhere:plugins/sana/skills/workday-agent-dialogue" into this agent's normal skills directory: use the OMGSkills MCP get_skill tool, install only the pinned_install commit and path, verify SKILL.md against pinned_install.skill_md_sha, and stop if MCP or the pin is unavailable. Do not overwrite or change existing skills without asking me.
Manual command
git clone https://github.com/Workday/everywhere /tmp/everywhere && ln -s /tmp/everywhere/plugins/sana/skills/workday-agent-dialogue ~/.claude/skills/workday-agent-dialogueFrom README
Workday agent dialogue Load workday first. Routing has already selected the agent. This skill does not choose an agent or an agentid, and it does not replace a matching catalog tool. A catalog tool returns on that call and does not use this skill. Batch agent questions An agent run commonly takes 30–90 seconds. Put every related part of the user's request into one complete message. Do not drip-feed questions that can be answered in one run. Tell the user once that Workday's agent is working and the answer takes about a minute. This batching rule does not turn separate operations into one request, and it does not override the gateway routing skill's fan-out or catalog-tool plan. Agent writes require explicit confirmation For a write routed to a Workday agent: Gather the complete intent before sending the first action. Relay the agent's proposed specifics and corrections to the user. Get the user's explicit confirmation.