supercli: Skill for Claude Code

.agents/skills/supercli_mcp_dev/SKILL.md

supercli-mcp-dev is a skill for Claude Code, Codex from javimosch/supercli. It costs 61 tokens per session (1,932 once invoked), scanned A, original, MIT.

A guide for developing Model Context Protocol (MCP) features in SuperCLI. MCP is a standard way for an application to connect to tools and services, and this guide covers its adapters, connection methods, streaming responses, and JSON-RPC messages.

In plain words
What is it for?
Use it when implementing MCP adapters, HTTP or server-sent event support, local process connections, proxying, or MCP server and client integration.
Why use it?
It gives developers the project-specific structure and message formats needed to add or change MCP connections without guessing how SuperCLI routes them.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: installed under .agents/ (shared by several agents).

This is javimosch/supercli's own configuration. It tells Claude Code and Codex how to work on supercli itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything supercli configures →

Needs its repository: it runs a file that does not travel with it, so clone the repository first. The line is node cli/mcp-daemon-client.js.

Reuse

Borrowing it

Nothing to install: this file belongs to javimosch/supercli. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/javimosch/supercli/master/.agents/skills/supercli_mcp_dev/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/javimosch/supercli

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 supercli-mcp-dev

README.md
[![agentmods](https://agentmods.dev/badge/skills/javimosch/supercli/supercli_mcp_dev/github.svg)](https://agentmods.dev/skills/javimosch/supercli/supercli_mcp_dev)
Your own site
<a href="https://agentmods.dev/skills/javimosch/supercli/supercli_mcp_dev"><img src="https://agentmods.dev/badge/skills/javimosch/supercli/supercli_mcp_dev/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 supercli-mcp-dev

Your own site · 80×15
<a href="https://agentmods.dev/skills/javimosch/supercli/supercli_mcp_dev"><img src="https://agentmods.dev/badge/skills/javimosch/supercli/supercli_mcp_dev.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 61 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,932 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. 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 $0.00061 $0.01932
Opus 5.5 $0.00024 $0.00773
Sonnet 5.5 $0.00012 $0.00386
Haiku 4.5 $0.00006 $0.00193

Measured today against content hash 475b96262850, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

supercli-mcp-dev scanned grade A 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 today.

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.

Makes network callslowCapability

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

curl http://localhost:3002/tool -X POST \
.agents/skills/supercli_mcp_dev/SKILL.md · 271 lines

How it starts

The opening of the file, as written. The whole thing — 271 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Supercli MCP Development

Work on MCP (Model Context Protocol) features in supercli including adapter implementation, SSE streaming support, and server integration.

MCP Adapter Architecture

The MCP adapter (cli/adapters/mcp.js) supports multiple transport modes:

Transport Modes

  1. HTTP/SSE MCP: HTTP-based MCP servers with SSE streaming
  2. Stdio MCP with Daemon: Local stdio MCP servers routed through daemon
  3. Stdio MCP Direct: Direct stdio calls (may have compatibility issues)

Key Components

  • execute(): Main entry point, routes to appropriate transport
  • callHttpMcpTool(): HTTP/SSE MCP handler with JSON-RPC support
  • parseSseResponse(): SSE event parser
  • callStdioTool(): Direct stdio execution
  • stdioCallToolJsonRpc(): Stdio JSON-RPC protocol handler
  • callDaemonTool(): Daemon-routed stdio calls

HTTP/SSE Implementation

Request Format

Uses JSON-RPC 2.0 format for HTTP MCP calls:

{
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "name": toolName,
    "arguments": input
  },
  "id": 1
}

Headers

Must include both content types for SSE negotiation:

{
  "Content-Type": "application/json",
  "Accept": "text/event-stream, application/json"
}

SSE Response Parsing

Parses Server-Sent Events format:

  • Lines starting with event: define event type
  • Lines starting with data: contain event data
  • Empty lines terminate events
  • Accumulates chunks and returns final result

Auto-detection

Automatically detects SSE responses via Content-Type header and falls back to JSON if not SSE.

Stdio MCP with Daemon

Configuration

{
  "adapterConfig": {
    "server": "mcp-server-name",
    "tool": "tool-name",
    "stateful": true,
    "timeout_ms": 60000
  }
}

Daemon Client

  • File: cli/mcp-daemon-client.js
  • Functions: callDaemonTool(), listDaemonTools()
  • Maintains persistent sessions for stateful MCP servers
  • Daemon runs on CLI side (not server) - starts on-demand when first command with stateful: true is executed via CLI
  • Runs as detached background process with Unix socket at ~/.supercli/mcp-daemon.sock
  • Persists across command executions to maintain MCP server sessions
  • Better for long-running stdio MCP servers

Read the full file on GitHub · 271 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. today First seen · 271 lines · 61 tokens per session scan A 475b96262850

Subscribe to this mod's changes

supercli-mcp-dev is a skill published in the GitHub repository javimosch/supercli (56 stars, last pushed today), licensed MIT. It adds 61 tokens to every session and 1,932 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-29.