model-delegation
Provides conventions for splitting development into high-level thinking sessions and mechanical execution subagents to optimize token usage.
Install
mkdir -p .claude/skills/model-delegation && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/19034" && unzip -o skill.zip -d .claude/skills/model-delegation && rm skill.zipInstalls to .claude/skills/model-delegation
Activation
This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.
Esposter model-delegation conventions — the main session does all thinking (specs, proposals, architecture, review); mechanical implementation is delegated to background subagents with self-contained prompts. Apply when deciding whether to implement in-session or delegate, and when writing a delegation prompt.Key capabilities
- →Delegate mechanical coding tasks to background subagents
- →Draft self-contained specs for subagent execution
- →Manage parallel PR batches using isolated git worktrees
- →Define verifiable done-definitions for automated tasks
- →Clean up agent worktrees and branches after PR merges
How it works
The main session performs high-level design and architecture, then offloads mechanical implementation to background subagents using self-contained prompts that include repo conventions, specs, and verification criteria.
Inputs & outputs
When to use model-delegation
- →Delegate mechanical refactors to background agents
- →Draft self-contained specs for subagent execution
- →Manage implementation boundaries between main sessions and subagents
- →Offload routine code migrations to background workers
About model-delegation
Defines a workflow for offloading repetitive coding tasks, such as migrations or mechanical refactors, to background agents. It instructs the user on how to craft self-contained delegation prompts so that subagents can execute tasks without requiring additional context from the main session.
Esposter model-delegation conventions — the main session does all thinking (specs, proposals, architecture, review); mechanical implementation is delegated to background subagents with self-contained prompts. Apply when deciding whether to implement in-session or delegate, and when writing a delegat
When not to use it
- →Exploratory or ideation work
- →Docs-authoring tasks
- →Tasks requiring judgment that compounds
Prerequisites
Limitations
- →Main session must not edit files staged by the agent
- →Two agents cannot run in the same git worktree
- →Never delete long-lived branches like nuxt5 or migrate-assets
How it compares
Unlike manual coding, this approach separates design thinking from mechanical execution by using isolated subagents to perform specific, well-scoped tasks.
Compared to similar skills
model-delegation side by side with the closest alternatives in the catalog.
| Skill | Installs | Updated | Safety | Difficulty |
|---|---|---|---|---|
| model-delegation (this skill) | 0 | 1mo | No flags | Advanced |
| tool-design | 3 | 3mo | Review | Intermediate |
| mcp-builder | 136 | 4mo | Review | Advanced |
| skill-creator | 128 | 4mo | Review | Advanced |
Try saying
Example prompts that trigger this skill in your AI assistant.
More by Esposter
View all by Esposter →You might also like
tool-design
muratcankoylan
This skill should be used when the user asks to "design agent tools", "create tool descriptions", "reduce tool complexity", "implement MCP tools", or mentions tool consolidation, architectural reduction, tool naming conventions, or agent-tool interfaces.
mcp-builder
anthropics
Guide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP) or Node/TypeScript (MCP SDK).
skill-creator
anthropics
Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends Claude's capabilities with specialized knowledge, workflows, or tool integrations.
skill-development
anthropics
This skill should be used when the user wants to "create a skill", "add a skill to plugin", "write a new skill", "improve skill description", "organize skill content", or needs guidance on skill structure, progressive disclosure, or skill development best practices for Claude Code plugins.
agent-identifier
anthropics
This skill should be used when the user asks to "create an agent", "add an agent", "write a subagent", "agent frontmatter", "when to use description", "agent examples", "agent tools", "agent colors", "autonomous agent", or needs guidance on agent structure, system prompts, triggering conditions, or agent development best practices for Claude Code plugins.
llama-factory
zechenzhangAGI
Expert guidance for fine-tuning LLMs with LLaMA-Factory - WebUI no-code, 100+ models, 2/3/4/5/6/8-bit QLoRA, multimodal support