Skip to content
Back to skills

Aped Zoom Out

ASecurity

Use when user says "zoom out", "step back", "take a step back", "broader view", "wider context", "lost the thread", "are we still solving the right problem", or invokes aped-zoom-out. Horizontal — invokable at any phase. Not for sprint progress dashboards (see aped-status) or for diff walks (see aped-checkpoint).

  • 2 stars
  • 0 votes
  • 0 copies
  • 2 views
  • Added May 27, 2026
developmentgobashgit

Works with

  • cli

Security analysis

A100/100

Pro scans all 2 files and shows the line behind each finding

Scanned May 27, 2026

npx -y skills add yabafre/aped-claude --skill aped-zoom-out --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Aped Zoom Out?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Aped Zoom Out
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/yabafre-aped-zoom-out/badge)](https://www.skillsdirectory.com/skills/yabafre-aped-zoom-out)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: aped-zoom-out
keep-coding-instructions: true
description: 'Use when user says "zoom out", "step back", "take a step back", "broader view", "wider context", "lost the thread", "are we still solving the right problem", or invokes aped-zoom-out. Horizontal — invokable at any phase. Not for sprint progress dashboards (see aped-status) or for diff walks (see aped-checkpoint).'
allowed-paths:
  write: ["{{OUTPUT_DIR}}/**", "{{APED_DIR}}/**"]
  read-only: ["src/**", "tests/**", "package.json"]
disable-model-invocation: true
license: MIT
metadata:
  author: yabafre
  version: {{CLI_VERSION}}
---
<!-- AUTO-GENERATED from SKILL.md.tmpl. Edits will be overwritten. Run: npm run gen:skill-docs -->

**Activation guard (6.2.0):** Before any other action, run `bash {{APED_DIR}}/scripts/check-enabled.sh`. If it exits non-zero, print "APED disabled — run aped-method enable" and HALT.

# APED Zoom Out — Step Back

Stop the current task. Look at the broader project context. Ask out loud: "Are we still solving the right problem?"

Read recent state (`{{OUTPUT_DIR}}/state.yaml`), recent decisions (top of `{{OUTPUT_DIR}}/lessons.md`, recent commits via `git log --oneline -20`), and the original goal (PRD, brief, ticket, or user's first message in this session). Compare what is being implemented now to that original goal. If they have drifted, surface the drift to the user *before* writing more code.

**Missing artefacts are signal, not error.** If `state.yaml` is absent, the project is pre-pipeline — say so and zoom out from `lessons.md` + git log + first user message only. If `lessons.md` is absent, say "no logged decisions" rather than inventing one. The four bullets below should always cite the source they drew from; bullets backed by *no* source are forbidden.

This skill produces no artefact. Its only output is the agent's re-orientation message to the user — a 4-bullet summary:
- **Where we started:** original goal, in one sentence.
- **Where we are:** what got done, in one sentence.
- **Drift, if any:** divergence between the two, with file/decision pointers.
- **Recommendation:** continue / re-anchor / stop and re-PRD.

Use this when you sense the implementation has been deep in details too long and the original framing is fading. Keep the answer to two sentences — brevity is the whole point.

## No State Change

Zoom-out is read-only. It does not advance pipeline state, does not write artefacts, does not modify state.yaml.

Files in this skill

  • SKILL.md2.4 KB
  • SKILL.md.tmpl2.2 KB

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.