mcpField
Z
zoom-out

zoom-out

github.com/akillnessDevelopment52275 installsUpdated today
Repository

About zoom-out

Zoom Out is a directive skill for coding agents that need a high-level map of an unfamiliar code area. It outlines relevant modules, caller and dependency relationships, and domain vocabulary before implementation work.

Key features

  1. 01Map relevant modules in an area
  2. 02Show caller relationships
  3. 03Identify key dependencies
  4. 04Apply domain vocabulary from CONTEXT.md

Use cases

  1. 01Understand how unfamiliar code fits the system
  2. 02Assess change impact before editing
  3. 03Build a module map before implementation

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/akillness/jeo-skills --skill zoom-out

After installing, type/zoom-outin Claude Code to confirm it loaded.

Details

From the project README

Zoom Out

Get a higher-level architectural perspective on unfamiliar code sections.

Upstream status

Removed upstream in mattpocock/skills (unused in practice). Kept here as the architecture zoom-out entry point.

When to use this skill

  • Navigating unfamiliar code and needing to understand how it connects to the larger system
  • Before making changes to understand the blast radius
  • When you need a module map before diving into implementation

When not to use this skill

  • Finding specific symbols or patterns → use codebase-search
  • Architectural improvement opportunities → use improve-codebase-architecture
  • Debugging a specific failure → use debugging

What zoom-out produces

A structural map showing:

  • All relevant modules in the area of interest
  • Caller relationships (what calls what)
  • Key dependencies (what depends on what)
  • Domain vocabulary from CONTEXT.md applied to module names

How to use

When you encounter unfamiliar code, ask for a zoom-out before diving in:

"Zoom out and give me a broader view of how [module/component] fits into the system."

The response will map the relevant modules using the project's domain glossary, not internal implementation names.

Output format

[Module A] → calls → [Module B], [Module C]
[Module B] ← called by → [Module A], [Module D]

Key seams:
- [Interface X] separates [Domain Concept 1] from [Domain Concept 2]
- [Interface Y] is where [Domain Concept 3] behavior can be swapped

Domain terms in use here:
- "Order" = [what it means in this context]
- "Customer" = [what it means in this context]

Note on invocation

This skill sets disable-model-invocation: true — it is a directive framework that instructs the agent how to respond, not an independent processing unit. Invoke it as a behavioral directive when exploring code.

Instructions

  1. Identify the task trigger and expected output.
  2. Follow the workflow steps in this skill from top to bottom.
  3. Validate outputs before moving to the next step.
  4. Capture blockers and fallback path if any step fails.

Examples

  • Example: Apply this skill to a small scope first, then scale to full scope after validation passes.

Best practices

  • Keep outputs deterministic and auditable.
  • Prefer small reversible changes over broad risky edits.
  • Record assumptions explicitly.

References

  • Project standards: .agent-skills/skill-standardization/SKILL.md
  • Validator script: .agent-skills/skill-standardization/scripts/validate_skill.sh

Comments

Markdown supported