director-mode

director-mode is a skill for Claude Code, Codex from claude-world/director-mode-lite. It costs 79 tokens per session (493 once invoked), scanned A, original, MIT.

A workflow for directing substantial coding work across Claude Code, Codex CLI, or Grok Build, which are command-line coding agents.

In plain words
What is it for?
Onboarding agents to a repository, splitting implementation and review work, planning verification, coordinating multiple agents, and handing work between coding tools.
Why use it?
It keeps a large task focused by defining the outcome, coordinating independent work, and requiring evidence before calling it complete.

Skill for Claude CodeCodex

Written for Claude Code and Codex: user-invocable in frontmatter, but also agents/openai.yaml present. Also seen: mentions Claude Code; mentions Codex.

Part of the director-mode-lite plugin — 36 skills, 14 agents shipped together

Good fit Onboarding agents to a repository, splitting implementation and review work, planning verification, coordinating multiple agents, and handing work between coding tools.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/claude-world/director-mode-lite/director-mode
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add claude-world/director-mode-lite --skill director-mode
Clone the repo
git clone --depth 1 https://github.com/claude-world/director-mode-lite

Made for: Claude Code, Codex.

Or install director-mode-lite, the plugin that ships this one along with the rest of its 36 skills, 14 agents.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for director-mode

README.md
[![agentmods](https://agentmods.dev/badge/skills/claude-world/director-mode-lite/director-mode/github.svg)](https://agentmods.dev/skills/claude-world/director-mode-lite/director-mode)
Your own site
<a href="https://agentmods.dev/skills/claude-world/director-mode-lite/director-mode"><img src="https://agentmods.dev/badge/skills/claude-world/director-mode-lite/director-mode/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for director-mode

Your own site · 80×15
<a href="https://agentmods.dev/skills/claude-world/director-mode-lite/director-mode"><img src="https://agentmods.dev/badge/skills/claude-world/director-mode-lite/director-mode.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 79 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 493 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector pass 7 Sept 2026
How audits are shown
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00079 $0.00493
Opus 5 $0.00039 $0.00246
Sonnet 5 $0.00016 $0.00099
Haiku 4.5 $0.00008 $0.00049

Measured 10d ago against content hash 1437b7e6ce02, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

director-mode scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 10d ago.

The scan reads SKILL.md. This mod also ships 1 executable file (scripts/director-doctor.py), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

skills/director-mode/SKILL.md · 50 lines

What it actually says

Director Mode

Use .director-mode/GUIDANCE.md when the project runtime is installed. For a plugin-only install, use this skill's bundled references/GUIDANCE.md instead. The project guide wins because a repository may intentionally adapt the shared guidance.

Keep the working contract small:

  1. State the outcome, relevant context, meaningful constraints, and completion evidence.
  2. Inspect the repository before choosing an approach.
  3. Use independent agents when parallel work or an independent review helps.
  4. Keep one integrator responsible for the final diff and verification.
  5. If another CLI should take over, use the session-relay skill.

Keep guidance proportional. A small request does not need a ceremony, a plan file, or an agent team. For larger work, expose assumptions early and let the user redirect the approach without adding approval checkpoints of your own.

Read-only diagnosis

When discovery, adapters, or hooks seem wrong, run director-doctor from the first available location:

  1. .director-mode/bin/director-doctor in the project;
  2. $HOME/.claude/bin/director-doctor from Bootstrap Kit;
  3. this skill's bundled scripts/director-doctor.py for plugin-only use.

Use --json --no-probe for deterministic inspection. The doctor reports known runtime, CLI, asset, and hook surfaces; it never repairs, installs, deletes, or changes permissions. Treat its recommendations as diagnostic leads, not gates.

Capability boundary

If the user explicitly chooses an unrestricted session in a trusted workspace, the installed director-open launcher can express that choice with the CLI's native controls. Do not launch it or rewrite persistent settings merely because this skill was selected.

These are operating suggestions, not a policy engine. Do not change the CLI's permissions, approvals, sandbox, network access, or hook trust on behalf of this skill.

Files

What ships with it

4 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 10d ago First seen · 50 lines · 79 tokens per session scan A 1437b7e6ce02

Subscribe to this mod's changes

director-mode is a skill published in the GitHub repository claude-world/director-mode-lite (81 stars, last pushed 10d ago), licensed MIT. It adds 79 tokens to every session and 493 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other skills, from other repositories

concise

Concise communication mode. Cuts 60-70% of output tokens while keeping natural, readable English. Drops filler, hedging, and pleasantries but maintains grammar and sentence flow. Elaborate on request -- ask for detail and get it, then auto-return to concise. Use when user says "concise mode", "be concise", "less…

o4f6bgpac3/concise · 92 tokens

peekaboo

Provides runtime observation and interaction for native macOS interfaces through accessibility state and screenshots. Use when the task depends on visible or interactive state in a running SwiftUI/AppKit app: what is rendered, focused, selected, enabled, reachable through menus/windows/dialogs, or experienced across a…

johnkozaris/jko-claude-plugins · 104 tokens

mobile-flows-maestro

This skill should be used when Maestro is explicitly requested or already present and the task is to author, run, or debug iOS/Android Maestro flows; use Maestro MCP; or handle Maestro selectors, system UI, permissions, Keychain, JavaScript, waits, device state, flakiness, or CI. Evidence includes a .maestro directory…

johnkozaris/jko-claude-plugins · 102 tokens

validate-mobile

Run a Maestro flow on an explicitly selected iOS or Android device and report behavioral evidence.

johnkozaris/jko-claude-plugins · 20 tokens

validate-api

Run the project's Hurl scenarios with an OIDC access token passed as a secret variable.

johnkozaris/jko-claude-plugins · 20 tokens

kanso-audit

Use when the user asks for a code review, audit, pre-PR check, quality sweep, or pattern analysis of a diff, branch, module, or codebase. Also use when the user asks to "check" or "look over" their code for issues.

blakecyze/kanso · 60 tokens