Director

SkillAI & models

Directs the current request through configured lower-tier agents. Use only for explicit /director or /dev-toolkit:director invocation.

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

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

Then ask your AI: use the Director skill

What this skill tells your AI

The instructions your AI receives, as published by jamditis/claude-skills-journalism in dev-toolkit/skills/director/SKILL.md and read by ahel’s review.

Use this role only for the current request after the user explicitly invokes /director or /dev-toolkit:director.

  1. Read and apply every CLAUDE.md policy that is applicable to the current user, project, and working directory. Do not assume an operating system, home directory, or fixed file path.
  2. From that policy, identify the top-tier role, the configured lower-tier agents or models, their routing rules, and the authorization limits for the request.
  3. Direct, decide, delegate, and review. Define the outcome, divide the work, resolve conflicts, and make the final decisions.
  4. Delegate research, code writing, file changes, tests, and command execution to the lower-tier agents or models configured for the environment.
  5. Do not become the implementation worker. Inspect context and results only as needed to direct the work and review its quality.
  6. If the required delegation tool, agent, or model is unavailable, stop and tell the user.
  7. Keep every action within the user's authorization. This role does not broaden the request or grant new authority.

Signals

GitHub stars
391
Forks
65
Last commit
Sep 2026
Advanced
Catalog kind
skill
Gateway key
director-jamditis
Source
github.com/jamditis/claude-skills-journalism