Stats
Actions
Tags
From bench
Generate a Laravel Action class — a single-purpose business operation with one `execute()` method and side effects (persistence, event dispatch, notifications). Use whenever the user describes a business operation like "create X", "send Y", "process Z", "mark W". For utility/calculator/parser classes use /service instead.
How this skill is triggered — by the user, by Claude, or both
Slash command
/bench:actionThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
You're the **/action** skill. Parse the user's request, ask one question if anything's ambiguous, delegate to the `action` agent, then synthesize its output.
You're the /action skill. Parse the user's request, ask one question if anything's ambiguous, delegate to the action agent, then synthesize its output.
The user's request: $ARGUMENTS
Extract:
{Verb}{Noun}Action (propose one if the user didn't name it)If the request describes a utility/calculator/parser (multiple methods, no orchestration side effects) → redirect to /service.
Otherwise: ask the user ONE question if anything's genuinely ambiguous. Don't ask three questions when one disambiguates the rest. For obvious cases, proceed.
Use the Task tool with subagent_type: "action". Pass the parsed args (class name, inputs, side effects, needs-current-user flag) — NOT the raw $ARGUMENTS. The agent reads the pattern, scaffolds, and reports back.
Re-frame the agent's output at the feature level for the user. Mention any follow-ups it flagged (missing event, missing model, suggested next skill).
Provides behavioral guidelines to reduce common LLM coding mistakes, focusing on simplicity, surgical changes, assumption surfacing, and verifiable success criteria.
Searches, retrieves, and installs Agent Skills from prompts.chat registry using MCP tools like search_skills and get_skill. Activates for finding skills, browsing catalogs, or extending Claude.
Creates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.
npx claudepluginhub pdx-apps/bench --plugin bench