conductor CLAUDE.md

Repository instructions for Conductor, an MCP server that gives AI agents tools through plugins. MCP is a standard way for AI tools to connect to external services.

In plain words
What is it for?
Building, testing, type-checking, linting, formatting, and running the Conductor command-line tool.
Why use it?
They give the coding assistant the project’s commands, architecture, testing routine, and build details in one place.

Instructions file

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 instructions/useconductor/conductor/claude-md
Clone the repo
git clone --depth 1 https://github.com/useconductor/conductor
Per session 1,333 This file is loaded in full into every session.
When invoked 1,333 The same file — it is already loaded in full.
Security scan C 2 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 $0.01333 $0.01333
Opus 5 $0.00666 $0.00666
Sonnet 5 $0.00267 $0.00267
Haiku 4.5 $0.00133 $0.00133

Measured 2d ago against content hash 7040330e8c96, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade C, and why

conductor CLAUDE.md scanned grade C with 2 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.

Downloads and executes remote codehighSupply chain

curl | sh runs whatever the server returns today, which is not necessarily what it returned when this was reviewed.

`install.sh` in root: one-line bash installer (`curl -fsSL .../install.sh | bash`). Checks Node ≥ 18, installs `@conductor/cli` globally, prompts to run `conductor init`.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

`install.sh` in root: one-line bash installer (`curl -fsSL .../install.sh | bash`). Checks Node ≥ 18, installs `@conductor/cli` globally, prompts to run `conductor init`.
CLAUDE.md · 119 lines

How it starts

The opening of the file, as written. The whole thing — 119 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CLAUDE.md

This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.

Commands

npm run build          # Compile TypeScript → dist/
npm run dev            # Watch mode (tsx watch)
npm test               # Run all tests (vitest)
npm run test:watch     # Watch mode tests
npm run test:coverage  # Coverage report
npm run typecheck      # Type-check without emit
npm run lint           # ESLint
npm run lint:fix       # ESLint with auto-fix
npm run format         # Prettier

Run a single test file:

npx vitest run tests/calculator.test.ts

Run the CLI locally (without building):

npx tsx src/cli/index.ts <command>

After building, run the CLI:

node dist/cli/index.js <command>
# or if globally linked:
conductor <command>

Tests live in tests/ and match **/*.test.ts. The postbuild script copies src/dashboard/index.html to dist/dashboard/index.html — if you add other static assets to src/dashboard/, add them to the postbuild script too.

Architecture

Conductor is an MCP (Model Context Protocol) server that exposes a plugin system as tools to AI agents. The codebase is ESM TypeScript ("type": "module", Node16 module resolution). All imports must use .js extensions even for .ts source files.

Core Object Model

Conductor (src/core/conductor.ts) is the central orchestrator. It owns:

  • ConfigManager — reads/writes ~/.conductor/config.json
  • DatabaseManager — SQLite via sql.js at ~/.conductor/conductor.db
  • PluginManager — lazy-initializes plugins on first use
  • AIManager — multi-provider AI (Claude, OpenAI, Gemini, Ollama)

Consumers should depend on the narrow interfaces in src/core/interfaces.ts (IConfig, IDatabase, IPluginRegistry) rather than the concrete classes.

Plugin System

Every capability is a plugin implementing the Plugin interface (src/plugins/manager.ts):

interface Plugin {
  name: string; description: string; version: string;
  initialize(conductor: Conductor): Promise<void>;
  isConfigured(): boolean;
  getTools(): PluginTool[];
  configSchema?: PluginConfigSchema;  // for `conductor config setup <plugin>`
  getContext?(): Promise<string | null>; // proactive reasoning cycle
}

Read the full file on GitHub · 119 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 · 119 lines · 1,333 tokens per session scan C 7040330e8c96

Subscribe to this mod's changes

conductor CLAUDE.md is an instructions file published in the GitHub repository useconductor/conductor (2 stars, last pushed 4mo ago), licensed Apache-2.0. It adds 1,333 tokens to every session, about $0.0067 per session on Opus 5. A static security scan graded it C with 2 findings (downloads and executes remote code, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.