gastown-orchestration

gastown-orchestration is a skill for Claude Code, Codex from jxoesneon/Ciel. It costs 41 tokens per session (1,159 once invoked), scanned A, original, Apache-2.0.

A command-line framework for coordinating multiple coding agents, tracking their work, and recovering after crashes.

In plain words
What is it for?
It is for starting and stopping workers, assigning batches of tasks, tracking task progress, and resuming interrupted workflows.
Why use it?
It organizes many agents and their tasks in one workflow, so work can continue when an agent or process stops unexpectedly.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: mentions subagents.

Good fit It is for starting and stopping workers, assigning batches of tasks, tracking task progress, and resuming interrupted workflows.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/jxoesneon/ciel/gastown-orchestration
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 jxoesneon/Ciel --skill gastown-orchestration
Clone the repo
git clone --depth 1 https://github.com/jxoesneon/Ciel

Made for: Claude Code, Codex.

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 gastown-orchestration

README.md
[![agentmods](https://agentmods.dev/badge/skills/jxoesneon/ciel/gastown-orchestration/github.svg)](https://agentmods.dev/skills/jxoesneon/ciel/gastown-orchestration)
Your own site
<a href="https://agentmods.dev/skills/jxoesneon/ciel/gastown-orchestration"><img src="https://agentmods.dev/badge/skills/jxoesneon/ciel/gastown-orchestration/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 gastown-orchestration

Your own site · 80×15
<a href="https://agentmods.dev/skills/jxoesneon/ciel/gastown-orchestration"><img src="https://agentmods.dev/badge/skills/jxoesneon/ciel/gastown-orchestration.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 41 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,159 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.
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 $0.00041 $0.01159
Opus 5 $0.00020 $0.00580
Sonnet 5 $0.00008 $0.00232
Haiku 4.5 $0.00004 $0.00116

Measured 8d ago against content hash dc0aa58b99b2, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

gastown-orchestration 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 8d 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.

skills/gastown-orchestration/SKILL.md · 71 lines

How it starts

The opening of the file, as written. The whole thing — 71 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CIEL ADAPTATION: Gas Town Orchestration (CLI Toolchain)

This skill adapts the Gas Town CLI toolchain (gt/bd) into Ciel's vocabulary. It provides a CLI-driven alternative to Ciel's internal fleet model — where autonomous-orchestration uses Ciel-native worktrees and missions, gastown uses external process spawning, bead-based work tracking, and mail-triggered patrols. Ciel's risk classifier gates subprocess spawning as mid→LLM_JUDGE or high→Council.

Core Vocabulary (Ciel Translation)

  • Bead (bd): A tracked work unit. Ciel equivalent: a mission task ticket.
  • Convoy: A batch of related beads. Ciel equivalent: a mission DAG node group.
  • Rig: A registered project workspace (GitHub repo). Ciel equivalent: an isolated worktree target.
  • Polecat: Ephemeral worker spawned per-task, vanishes on completion. Ciel equivalent: a one-shot sub-agent.
  • Crew: Persistent named worker with ongoing sessions. Ciel equivalent: a long-lived fleet agent.
  • Hook: Where work lands for a worker. GUPP (Gas Town Universal Propulsion Principle): if hook has work, run it.
  • Molecule: A workflow unit that survives crashes — any worker can resume where another left off.
  • Patrols: Witness (watches for stuck workers), Refinery (merges completed work), Mayor (coordinates rigs).

CLI Command Patterns

Engine Control: gt up (start), gt down (graceful stop), gt status (overview) Work Management: gt sling <bead> <rig> (assign work), gt convoy list, gt hook Workers: gt polecat list, gt crew list, gt peek <agent>, gt nudge <agent> "msg" Diagnostics: gt doctor (health check), gt doctor --fix (auto-repair), bd doctor, gt feed (activity stream) Beads: bd list, bd show <id>, bd sync, bd create --title "..." Refinery: gt refinery start, gt refinery status, gt refinery queue Patrol Activation: gt mail send <rig>/witness -s "Patrol" -m "Process completed work"

Lifecycle Management

  1. Install: go install .../gt@latest + go install .../bd@latest; verify with gt doctor AND bd doctor.
  2. Add Rig: Register a project; verify patrols exist via gt doctor --fix.
  3. Create Work: bd create --title "..." → bead ID assigned.
  4. Sling: gt sling <bead> <rig> → polecat spawns, work lands on hook, GUPP executes.
  5. Monitor: gt peek <agent>, gt feed; Witness patrols detect stuck workers.
  6. Merge: Refinery processes completed work → code lands on main.
  7. Shutdown: gt down for graceful stop.

Read the full file on GitHub · 71 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. 8d ago First seen · 71 lines · 41 tokens per session scan A dc0aa58b99b2

Subscribe to this mod's changes

gastown-orchestration is a skill published in the GitHub repository jxoesneon/Ciel (1 stars, last pushed 4d ago), licensed Apache-2.0. It adds 41 tokens to every session and 1,159 once invoked, about $0.0002 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 2026-09-03.

Related

Other skills, from other repositories