conjure

conjure is a command for coding agents from john-farina/claude-plugins. It costs 62 tokens per session (2,578 once invoked), scanned B, original, MIT.

A general Conjure command that turns a plain-language request into a Claude Code skill, agent, automatic event script, or connection to an outside service.

In plain words
What is it for?
Use it when you want to describe an extension in ordinary language and have Conjure choose, question, and build the appropriate form.
Why use it?
It removes the need to know the extension type or write its setup files by hand.

Command

Part of the conjure plugin — 7 commands shipped together

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 commands/john-farina/claude-plugins/conjure
Clone the repo
git clone --depth 1 https://github.com/john-farina/claude-plugins

Or install conjure, the plugin that ships this one along with the rest of its 7 commands.

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 conjure

README.md
[![agentmods](https://agentmods.dev/badge/commands/john-farina/claude-plugins/conjure.svg)](https://agentmods.dev/commands/john-farina/claude-plugins/conjure)
Your own site
<a href="https://agentmods.dev/commands/john-farina/claude-plugins/conjure"><img src="https://agentmods.dev/badge/commands/john-farina/claude-plugins/conjure.svg" alt="Measured on agentmods" height="20"></a>
Per session 62 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,578 The whole file, excluding the scripts and references it only reads on demand.
Security scan B 1 finding. 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 $0.00062 $0.02578
Opus 5 $0.00031 $0.01289
Sonnet 5 $0.00012 $0.00516
Haiku 4.5 $0.00006 $0.00258

Measured 3d ago against content hash 11e5116f9d07, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade B, and why

conjure scanned grade B with 1 finding 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 3d 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.

Reads agent configuration directoriesmediumAgent snooping

.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.

cat ~/.claude/conjure-config.md 2>/dev/null
plugins/conjure/commands/conjure.md · 237 lines

How it starts

The opening of the file, as written. The whole thing — 237 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Phase 0: Help Check

If $ARGUMENTS is "help" or "--help" or "-h": tell the user to run /conjure-help for the full reference, then stop.


Phase 0.5: Load User Preferences

Read both conjure config files if they exist:

cat ~/.claude/conjure-config.md 2>/dev/null
cat .claude/conjure-config.md 2>/dev/null

Follow all instructions found in ## global and any type-specific sections throughout this command. Project-local instructions (.claude/conjure-config.md) take precedence over global ones when they conflict. If neither file exists, proceed with defaults.


Phase 1: Classify + Initial Intake

The user wants to build something. They may or may not know what type.

Their request: $ARGUMENTS

Run the classification + intent batch. Use AskUserQuestion with these questions (skip any that $ARGUMENTS already answers clearly):

Q1 "Access" — "Does this need to connect Claude to an external system it can't reach natively (a database, REST API, third-party service like Jira/Linear/Notion)?" Options:

  • Yes — Claude needs live access to something external
  • No — it works with what Claude already has access to
  • Not sure

Q2 "Trigger" — "How does this get activated?" (skip if Q1 = Yes) Options:

  • I type a /command in the prompt
  • Automatically when something happens (file edit, tool call, session start, stop)
  • Claude spawns it programmatically with its own context and tools
  • Not sure — I just know what I want it to do

Q3 "Job" — "What is the ONE thing this does? Be specific — not 'helps with X' but 'does Y when Z happens.'" Options:

  • Connects Claude to an external system / API / database
  • Runs a multi-step workflow I trigger manually
  • Fires automatically and blocks, logs, or injects context
  • Executes a focused research or analysis task in isolation

Q4 "Scope" — "Should this work in all projects or just this one?" Options:

  • All projects (user-global)
  • This project only
  • Not sure

Phase 2: Classify

Read the full file on GitHub · 237 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. 3d ago First seen · 237 lines · 62 tokens per session scan B 11e5116f9d07

Subscribe to this mod's changes

conjure is a command published in the GitHub repository john-farina/claude-plugins (1 stars, last pushed 3mo ago), licensed MIT. It adds 62 tokens to every session and 2,578 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it B with 1 finding (reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.