agents opus

Prompt-driven

Give a free-form prompt; the skill auto-selects a mode and tells you why. No keyword grammar.

Quick reference

FieldValue
Command/brewcode:agents
Inputfree-form prompt or empty
Specialistbrewcode:agent-creator
List source.claude/agents/, ~/.claude/agents/, plugin agents/
Modelopus

When to use

  • Check the status of all agents across project and global scopes
  • Create a new agent from a plain-English description
  • Improve an existing agent’s trigger phrases or system prompt
  • Review an agent for quality, focus, and trigger coverage
  • Batch-create or batch-improve a set of agents in one pass

Examples

/brewcode:agents
/brewcode:agents status of my agents
/brewcode:agents create a backend validator agent
/brewcode:agents improve reviewer

How it works

  1. Input gate

    Reads $ARGUMENTS. An empty prompt skips analysis and jumps directly to the No-prompt menu; a non-empty prompt proceeds to mode selection.

  2. Auto-mode select

    Announces Mode: <mode> (agents) — chosen because … based on keywords, intent, and target presence in the prompt.

  3. No-prompt menu

    Sends a single AskUserQuestion with options: Status (agents) (recommended), Status (all: agents+rules+skills), Create, Improve, Review, List (plain), Cancel. The skill continues with the chosen mode.

  4. Dispatch

    Routes to the appropriate handler. Create and improve modes delegate to the brewcode:agent-creator specialist via Task tool, passing full context — description, scope, model, and improvement focus.

  5. Real status

    Scans .claude/agents/, ~/.claude/agents/, and plugin agents/ directories; reads each agent file to extract name, model, and trigger count. Reports scope, path, and quality signal per agent.

  6. Final output

    Prints a detection + result table: mode chosen, target, file path, model, scope, and CLAUDE.md update status. Suggests next steps such as /reload-plugins and a test invocation phrase.

Modes

ModeTriggerAction
statusdefault — no explicit mode keywordScans all scopes, prints agents with model, scope, and trigger count
listlist keywordGlobs *.md across ALL agent scopes (.claude/agents/, ~/.claude/agents/, brewcode/agents/)
createcreate <description>Scaffolds a new agent file via agent-creator; asks scope, model, CLAUDE.md update
improveimprove <name> or bare <name|path>Tunes description triggers and system prompt quality
reviewreview <name>Full quality review: trigger coverage, prompt focus, redundancy check
batchauto-detected fan-out flagNot a menu item — auto-detected from prompt (“batch”, “all”, “each”); triggers parallel delegation when creating or improving multiple agents
Create / improve parameters
ParameterOptionsNotes
ScopeProject / Global / PluginProject → .claude/agents/; Global → ~/.claude/agents/; Plugin → brewcode/agents/
Modelsonnet / opus / haiku / inheritinherit keeps the model from the existing file (improve only)
CLAUDE.md updateyes / noAdds or updates the agent row in the agents table; appends a new section if none exists
Improve focustriggers / system prompt / both / full reviewFor improve and review modes
Descriptionfree-form, aim ≤100 charsUsed as the agent description frontmatter — drives LLM auto-invocation
Specialistbrewcode:agent-creatorHandles file scaffolding and prompt engineering
📝

rules

Sibling: manage project rules
✏️

skills

Sibling: manage skills
🚀

brewcode

Plugin overview
🔗

GitHub source

SKILL.md

Updating plugins

Use /brewtools:plugin-update to check and update the brewcode plugin suite in one command. See the FAQ for details.