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.
npx agentmods add commands/john-farina/claude-plugins/conjure-mcpgit clone --depth 1 https://github.com/john-farina/claude-pluginsWrote 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.
[](https://agentmods.dev/commands/john-farina/claude-plugins/conjure-mcp)<a href="https://agentmods.dev/commands/john-farina/claude-plugins/conjure-mcp"><img src="https://agentmods.dev/badge/commands/john-farina/claude-plugins/conjure-mcp.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00035 | $0.02121 |
| Opus 5 | $0.00017 | $0.01060 |
| Sonnet 5 | $0.00007 | $0.00424 |
| Haiku 4.5 | $0.00003 | $0.00212 |
Grade B, and why
conjure-mcp 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 5d 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 How it starts
The opening of the file, as written. The whole thing — 252 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Phase 0: Route the Request
Before building anything, verify an MCP server is the right artifact.
| Question | If YES → use instead |
|---|---|
| Does Claude already have access to this data/system via native tools? | Skill → run /conjure-skill <name> |
| Is this a repeatable workflow or checklist, not a live connection? | Skill → run /conjure-skill <name> |
| Should this fire automatically on a tool call event? | Hook → run /conjure-hook <description> |
| Does Claude spawn this with isolated context? | Agent → run /conjure-agent <description> |
| Does this require a live, typed, schema-validated connection to an external system? | MCP → continue ✅ |
MCP signal phrases — if you hear any of these, it's an MCP:
| Request sounds like… | Actually is… |
|---|---|
| "connect to / integrate with / access [external system]" | MCP ✅ |
| "query our database / hit our REST API / read from Postgres" | MCP ✅ |
| "stop copy-pasting from Jira/Linear/Figma/Slack" | MCP ✅ |
| "Claude needs to call [external service]" | MCP ✅ |
| "hook that fetches from our API" | MCP — hooks can't make authenticated external calls reliably |
| "agent to query the database" | MCP — agents use tools; the tool should be an MCP server |
| "skill to read from Notion" | MCP — skills are procedural, not connective |
Cost trade-off: Each MCP server with ~10 tools consumes ~10,000 tokens at session start (always). A skill costs ~100 tokens until activated. Only use MCP when the external access is genuinely required.
Phase 0.5: Load User Preferences
Before doing anything else, 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 ## mcp and ## global sections throughout every phase of this command. Project-local instructions (.claude/conjure-config.md) take precedence over global ones when they conflict. If neither file exists, proceed with defaults.
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.
- 5d ago First seen · 252 lines · 35 tokens per session scan B 0ea020564797
conjure-mcp is a command published in the GitHub repository john-farina/claude-plugins (1 stars, last pushed 3mo ago), licensed MIT. It adds 35 tokens to every session and 2,121 once invoked, about $0.0002 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.
Other commands, from other repositories
strip
This is the task-based stripper, not the always-on prior: a deliberate cleanup pass you asked for. Apply the fp-minify doctrine to the target and remove or compress comments that don't earn their place.
web-design
Design and build a complete website or webapp from a project brief — end-to-end workflow from design to working React/Vite code.
design
Create a design direction and styleguide for a project — produces aesthetic direction, fonts, colors, imagery, and media prompts.
media-generate
Generate a media asset — image, video, music, or speech — with guided prompts and sensible defaults.
media-assets
List, organize, and manage generated media assets in the output directory.
aws
Structured guidance for AWS CloudFront distributions, WAF web ACLs, Lambda@Edge, CloudFront Functions, Firewall Manager multi-account enforcement, and IAM/IRSA patterns. Covers OAC, cache policies, security headers, managed rule groups, rate limiting, FMS FIRST/MIDDLE/LAST ownership model, and production-ready…