Custom-MCPs: Skill for Claude Code

.agents/skills/add-fleet-connector/SKILL.md

add-fleet-connector is a skill for Claude Code, Codex from Edison-Watch/Custom-MCPs. It costs 50 tokens per session (1,422 once invoked), scanned A, original, MIT.

A skill for adding an HTTP-based MCP server to the Edison marketplace, where an MCP server connects an agent to an external service.

In plain words
What is it for?
Use it to scaffold a fleet connector, describe its tools, classify each tool's access to data and ability to change things, and run the marketplace checks.
Why use it?
It guides the required catalog entry, icon, tool safety classifications, and validation checks so the connector is not rejected or treated as unsafe by default.

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 Edison-Watch/Custom-MCPs's own configuration. It tells Claude Code and Codex how to work on Custom-MCPs 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 Custom-MCPs configures →

Needs its repository: it reads a path above its own folder, which exists only inside the repository. The line is export * from "../../../shared/auth/ts/auth";.

Reuse

Borrowing it

Nothing to install: this file belongs to Edison-Watch/Custom-MCPs. 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/Edison-Watch/Custom-MCPs/main/.agents/skills/add-fleet-connector/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/Edison-Watch/Custom-MCPs

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 add-fleet-connector

README.md
[![agentmods](https://agentmods.dev/badge/skills/edison-watch/custom-mcps/add-fleet-connector/github.svg)](https://agentmods.dev/skills/edison-watch/custom-mcps/add-fleet-connector)
Your own site
<a href="https://agentmods.dev/skills/edison-watch/custom-mcps/add-fleet-connector"><img src="https://agentmods.dev/badge/skills/edison-watch/custom-mcps/add-fleet-connector/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 add-fleet-connector

Your own site · 80×15
<a href="https://agentmods.dev/skills/edison-watch/custom-mcps/add-fleet-connector"><img src="https://agentmods.dev/badge/skills/edison-watch/custom-mcps/add-fleet-connector.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 50 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,422 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 $0.00050 $0.01422
Opus 5 $0.00025 $0.00711
Sonnet 5 $0.00010 $0.00284
Haiku 4.5 $0.00005 $0.00142

Measured 6d ago against content hash e05f60a18960, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

add-fleet-connector 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 6d 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.

.agents/skills/add-fleet-connector/SKILL.md · 132 lines

How it starts

The opening of the file, as written. The whole thing — 132 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Adding a fleet MCP connector

A fleet connector is one HTTP-wrapper MCP server advertised to the Edison marketplace. Its source of truth is servers/<id>/catalog-entry.json (+ a co-located <id>.svg icon) in this repo; edison-watch mirrors it downstream. Full contract: shared/catalog/README.md.

The one step you cannot skip is classifying every tool the server exposes. Marketplace installs skip autoconfig auto-labeling, so a tool with no classification mounts at the protective default (write + read_private + read_untrusted + SECRET) and trips the lethal-trifecta guard on its first call - the connector looks broken. make catalog_check fails until every edison_hosted connector ships a non-empty tools_configurations.

1. Scaffold

make new-connector id=<id>          # id: lowercase letters, digits, hyphens

This writes servers/<id>/catalog-entry.json (a skeleton with TODOs, no tools_configurations yet) and a placeholder servers/<id>/<id>.svg.

The scaffold covers the catalog entry only. You still hand-write the Worker itself - src/index.ts (the MCP tool over streamable HTTP), wrangler.jsonc, and test/ - mirroring an existing connector such as servers/reddit.

Auth is shared - re-export it, never copy it. The verify layer (open | bearer | edison-jwt with JWKS verification) lives once in shared/auth/ts. Make src/auth.ts and src/jwt.ts one-line re-exports:

// src/auth.ts
export * from "../../../shared/auth/ts/auth";
// src/jwt.ts
export * from "../../../shared/auth/ts/jwt";

Never copy another connector's auth.ts / jwt.ts: it forks security-sensitive code into an independently-maintained copy, so a bug fixed in one silently persists in the others.

When you write the tool handler, validate and bound the inputs a PUBLIC caller can send: cap array lengths, verify or allow-list URLs (reject blank, off-domain, or credential-bearing ones), and reject silently-conflicting arguments. An unbounded input forwarded to a paid upstream is an abuse and cost vector.

Read the full file on GitHub · 132 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. 6d ago Changed · +33 lines e05f60a18960
  2. 9d ago First seen · 99 lines · 50 tokens per session scan A 7146888ef588

Subscribe to this mod's changes

add-fleet-connector is a skill published in the GitHub repository Edison-Watch/Custom-MCPs (0 stars, last pushed yesterday), licensed MIT. It adds 50 tokens to every session and 1,422 once invoked, about $0.0003 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 2026-08-31.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

local-ai-agents

Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…

microsoft/ai-agents-for-beginners · 200 tokens

next-cache-components-adoption

Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…

vercel/next.js · 95 tokens

insight-error-page

Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…

vercel/next.js · 83 tokens

next-cache-components-optimizer

Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…

vercel/next.js · 170 tokens

next-partial-prefetching-adoption

Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…

vercel/next.js · 103 tokens