sandbox-activate

SkillAI & models

Activate a sandbox item by symlinking it to the Claude Code discovery location

Available today. Use it from your connected AI after setup.

No other account needed.

Connect ahel once, and every AI you use reads what you have installed.

Then ask your AI: use the sandbox-activate skill

What this skill tells your AI

The instructions your AI receives, as published by the-agency-ai/the-agency in .claude/skills/sandbox-activate/SKILL.md and read by ahel’s review.

Wire up a sandbox item so Claude Code discovers and uses it.

$ARGUMENTS is <name> — the item name (without extension).

Detect engineer from usr/*/ directories.
Find the item by searching usr/<engineer>/ for files matching <name>:
- usr/<engineer>/claude/commands/<name>.md → command
- usr/<engineer>/agency/hooks/<name>.sh → hook
- usr/<engineer>/agency/hookify/<name>.md → hookify
- usr/<engineer>/tools/<name>.* → tool
- usr/<engineer>/scripts/<name>.* → script
Check for existing symlink. If one exists, show target and ask to replace.
Create symlink based on type:
- command: ln -sf ../../usr/<engineer>/claude/commands/<name>.md .claude/commands/usr-<engineer>.<name>.md
- hookify: ln -sf ../usr/<engineer>/agency/hookify/<name>.md .claude/hookify.usr-<engineer>.<name>.local.md
- hook: create symlink + print the settings.local.json entry needed for manual wiring
- tool/script: print PATH instruction
Verify the symlink resolves.
Print confirmation.

Advanced