conductor

A project-planning skill that maintains Conductor-compatible context, specifications, implementation plans, and progress records.

In plain words
What is it for?
Setting up project context, starting or resuming tracks, implementing planned work, checking progress, and reverting changes.
Why use it?
It keeps project planning files in the structure expected by the Conductor workflow and Gemini Conductor extension.

Skill for Claude CodeCodex

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.

agentmods
npx agentmods add skills/pythrick/codex-conductor/conductor
Any agent
npx skills add pythrick/codex-conductor --skill conductor
Clone the repo
git clone --depth 1 https://github.com/pythrick/codex-conductor

Made for: Claude Code, Codex.

Per session 17 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,294 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00017 $0.01294
Opus 5 $0.00009 $0.00647
Sonnet 5 $0.00003 $0.00259
Haiku 4.5 $0.00002 $0.00129

Measured 2d ago against content hash eb4d9403b67d, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

conductor 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 2d ago.

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.

.codex/skills/conductor/SKILL.md · 129 lines

How it starts

The opening of the file, as written. The whole thing — 129 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Conductor Compatibility Skill

Purpose: create and maintain Conductor artifacts so they remain fully compatible with the Gemini Conductor extension and its workflow (Context -> Spec & Plan -> Implement).

Use this skill when the user asks to set up a new project context, start a new track, implement or resume a track, check progress, or revert work.

Required File Structure (Do Not Deviate)

conductor/
  product.md
  product-guidelines.md
  tech-stack.md
  workflow.md
  setup_state.json
  code_styleguides/
    *.md
  tracks.md
  tracks/
    <track_id>/
      spec.md
      plan.md
      metadata.json
  archive/                 (optional)
    <track_id>/            (optional)

Required Formats

Track ID

  • Format: shortname_YYYYMMDD (lowercase ASCII, underscores only).
  • Example: auth_20251219.

conductor/tracks.md

  • Must use --- separators between track sections.
  • Each track section must use the exact heading format below.
---

## [ ] Track: <Track Description>
*Link: [./conductor/tracks/<track_id>/](./conductor/tracks/<track_id>/)*

conductor/tracks/<track_id>/metadata.json

  • Required keys: track_id, type, status, created_at, updated_at, description.
  • Timestamps are ISO 8601 UTC, e.g., 2025-12-19T23:59:59Z.
  • status values: new, in_progress, completed, cancelled.

conductor/tracks/<track_id>/plan.md

  • Use checklist markers: [ ] (pending), [~] (in progress), [x] (done).
  • When completing a task, append the first 7 chars of the commit SHA to the task line.
  • If the workflow defines the Phase Completion Verification protocol, append a meta-task to each phase: - [ ] Task: Conductor - User Manual Verification '<Phase Name>' (Protocol in workflow.md)
  • When a phase is checkpointed, append [checkpoint: <sha>] to the phase heading.

Setup Workflow (Greenfield or Brownfield)

  1. Detect project maturity (greenfield vs brownfield) based on repo and code presence.
  2. Create conductor/ and conductor/setup_state.json if missing. Start with: { "last_successful_step": "" }
  3. Generate or update the core documents with user input:
    • product.md
    • product-guidelines.md
    • tech-stack.md
  4. Copy the workflow template from assets/templates/workflow.md into conductor/workflow.md unless the user provides a custom workflow.
  5. Copy code style guides from assets/templates/code_styleguides/ into conductor/code_styleguides/.
  6. Initialize conductor/tracks.md using the format in assets/templates/tracks.md.
  7. Create the initial track folder and artifacts (spec.md, plan.md, metadata.json).
  8. Update conductor/setup_state.json as steps complete using these values:
    • 2.1_product_guide
    • 2.2_product_guidelines
    • 2.3_tech_stack
    • 2.4_code_styleguides
    • 2.5_workflow
    • 3.3_initial_track_generated

Read the full file on GitHub · 129 lines

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. 2d ago First seen · 129 lines · 17 tokens per session scan A eb4d9403b67d

Subscribe to this mod's changes

conductor is a skill published in the GitHub repository pythrick/codex-conductor (5 stars, last pushed 8mo ago), licensed Apache-2.0. It adds 17 tokens to every session and 1,294 once invoked, about $0.0001 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 2026-08-31.