Trainer Agent — Development agent for Claude Code
Improvement Proposal Agent of the Governed Multi-Agent Delivery System.
How to install Trainer Agent
Installs to ~/.claude/agents/ricardosantos0-claudius-trainer_agent.md
mkdir -p ~/.claude/agents && curl -fsSL https://raw.githubusercontent.com/RicardoSantos0/claudius/HEAD/agents/trainer_agent.md -o ~/.claude/agents/ricardosantos0-claudius-trainer_agent.md Restart Claude Code, or start a new session, for it to be picked up.
What Trainer Agent does
name: trainer-agent description: "Improvement Proposal Agent of the Governed Multi-Agent Delivery System. Invoked after every evaluation cycle. Reads evaluation reports, identifies improvement patterns, and produces advisory proposals for agents, policies, and workflows. Authority Level: L0 advisory — proposes only, never applies changes. All proposals require Master Orchestrator approval." tools: Read, Grep, Glob, TodoWrite model: inherit model_profile: auto
You are the **Trainer Agent
Alternatives in Development
- Dredd — Independent, on-demand AL codebase auditor for Business Central 109 ★
- Lite Drafter — Produces a short, triage-grade paper summary — Key Takeaways, Background, Main Idea & Summary, Critique 95 ★
- Rust Import Cleaner — Use this agent when the user wants to remove unused imports from Rust code to eliminate compiler warnings 77 ★
Full documentation available on GitHub
View Source RepositoryRelated Agents
Spawner Agent
Agent Designer of the Governed Multi-Agent Delivery System. Invoked only after Master Orchestrator approves a
Retrospective Agent
Daily / weekly / monthly self-improvement retrospective. Reads the auto-generated dev-log digests (docs/dev-lo
Claude Md Reviewer
Reviews CLAUDE.md improvement proposals against the existing codebase and CLAUDE.md files, producing tightly s
Kordal Investigator
Investigates one improvement question about a codebase and returns evidenced findings and ranked proposals. Di
Memory Proposal Collector
Reference documentation (NOT a dispatchable agent) for the coordinator-direct AUQ rendering flow at session-en
Disruptor
Controlled-wildness proposals for flat chapters; proposal-only.
Related Skills
Skill Evolution
Self-improvement feedback loop for AI agent skills — analyzes past sessions, drafts structured improvement pro
House Party Protocol
Evidence-first harness for coding agents across Claude Code and Codex CLI: cross-model review, governed loops,
Proposal Tick
Run proposal refine/review/deep-lit loops in parallel until proposals are ready.