Aux Delegator — Project Management agent for Claude Code
Invokes external CLI coding agents (Codex, OpenCode, Claude -p, ACP).
How to install Aux Delegator
Installs to ~/.claude/agents/dcstolf-speccraft-aux-delegator.md
mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/DCSTOLF/speccraft/HEAD/agents/aux-delegator.md -o ~/.claude/agents/dcstolf-speccraft-aux-delegator.md Restart Claude Code, or start a new session, for it to be picked up.
What Aux Delegator does
name: aux-delegator description: "Invokes external CLI coding agents (Codex, OpenCode, Claude -p, ACP). Use whenever a task should be offloaded to a non-Claude-Code model for parallelism, second opinion, or cost reasons." tools: [Bash, Read] model: sonnet
You are the aux-delegator. Your job is to take a task + context bundle and shell out to the requested aux agent, then return its output cleanly.
Inputs you receive
agent_name(must exist in.speccraft/agents.toml)task: t
Alternatives in Project Management
- Agent Evaluator — Evaluates agent output against 5-axis quality rubric (accuracy, completeness, clarity, actionability, concisen 243.5k ★
- Search Conversations — Use when searching Claude Code and Codex history for information learned in past conversations or prior experi 469 ★
- Codex Artist — Image-generation subagent 249 ★
Full documentation available on GitHub
View Source RepositoryRelated Agents
Arthur
Independent code review via the Codex CLI (read-only, headless) for a second opinion that reads project files
Codex Analyst
Token-heavy READ-ONLY analysis, codebase surveys, audits and inventories, executed by the Codex CLI rather tha
Envoy Runner
Proactively use when Claude Code is stuck, wants a second implementation or diagnosis pass, needs a deeper roo
Kimi Worker
Forward a coding, diagnosis, or review task to Kimi K3 (Moonshot) through the local kimi-worker.sh runtime. Us
Bfl Node Reviewer
Independent second-opinion diff reviewer for ComfyUI-FLUX-BFL-API. Loads BFL invariants into fresh context and
Mp Adversarial Reviewer
Adversarial second-opinion review of a completed masterplan task. The harness runs this agent on the routing p
Related Skills
Dsh Auto Review
Second-model AI auto-review for DeepSeek Harness approval requests: a read-only reviewer subagent returns stru
Second Opinion
Trail of Bits security skill for second opinion
Ask GPT
Get GPT (Codex) second opinion on a question or current work. Single-shot, advisory, no contamination.