losant-mcp-server: Skill for Claude Code

.claude/skills/mcp-add-tool-guide/SKILL.md

mcp-add-tool-guide is a skill for Claude Code from Losant/losant-mcp-server. It costs 70 tokens per session (1,902 once invoked), scanned A, original, MIT.

A guide for creating and registering documentation for a new MCP tool. MCP, or Model Context Protocol, is a standard way for an AI application to call tools and read resources.

In plain words
What is it for?
Use it when adding a tool under the specified source directory or updating its companion guide resource.
Why use it?
It keeps the guide's parameter details and resource tables aligned with the tool's actual code instead of relying on manually maintained documentation. It also provides a standard structure for existing guides.

Skill for Claude Code

Written for Claude Code: installed under .claude/.

This is Losant/losant-mcp-server's own configuration. It tells Claude Code how to work on losant-mcp-server 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 losant-mcp-server configures →

Reuse

Borrowing it

Nothing to install: this file belongs to Losant/losant-mcp-server. 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/Losant/losant-mcp-server/develop/.claude/skills/mcp-add-tool-guide/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/Losant/losant-mcp-server

Made for: Claude Code.

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 mcp-add-tool-guide

README.md
[![agentmods](https://agentmods.dev/badge/skills/losant/losant-mcp-server/mcp-add-tool-guide/github.svg)](https://agentmods.dev/skills/losant/losant-mcp-server/mcp-add-tool-guide)
Your own site
<a href="https://agentmods.dev/skills/losant/losant-mcp-server/mcp-add-tool-guide"><img src="https://agentmods.dev/badge/skills/losant/losant-mcp-server/mcp-add-tool-guide/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 mcp-add-tool-guide

Your own site · 80×15
<a href="https://agentmods.dev/skills/losant/losant-mcp-server/mcp-add-tool-guide"><img src="https://agentmods.dev/badge/skills/losant/losant-mcp-server/mcp-add-tool-guide.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 70 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,902 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 $0.00070 $0.01902
Opus 5.5 $0.00028 $0.00761
Sonnet 5.5 $0.00014 $0.00380
Haiku 4.5 $0.00007 $0.00190

Measured 16d ago against content hash 5ed7fd7ce66a, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

mcp-add-tool-guide 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 16d 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.

.claude/skills/mcp-add-tool-guide/SKILL.md · 108 lines

How it starts

The opening of the file, as written. The whole thing — 108 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Add an MCP Tool Guide

Every losant_* tool has a companion guide resource served at losant://guides/tools/{toolName}. The tool's own description points at it ("Read losant://guides/tools/losant_x before using this tool"), so the guide — not the tool description — is where the procedure, parameter semantics, and resource-type tables live.

A guide is a JS module that builds one markdown string at module load from the maps in src/constants.js, then exports it. Tables are generated, never hand-written, so they cannot drift from the code.

1. Gather the facts first

Before writing anything, read:

  • src/mcp/tools/<tool>.js — the real inputSchema, validation rules, and defaults. Every row of the Parameters table must match this file: which params are required, which are conditional and on what, which are rejected for which operations, what gets silently overwritten.
  • src/constants.js — the resource lists and maps the tables are built from.
  • An existing guide as the structural model: src/mcp/resources/tools/write-tool-guide.js (schema-driven columns) or delete-tool-guide.js (supported/unsupported column).

Do not describe behavior the tool does not have. If the guide would claim the tool errors on something, verify it in the tool source; document what actually happens, including silent coercions.

2. Create the file

Directory: src/mcp/resources/tools/

Filename: take toolName, strip the losant_ prefix, replace every _ with -, append -guide.js.

toolName file
losant_query query-tool-guide.js
losant_timeseries timeseries-tool-guide.js
losant_device_state device-state-tool-guide.js

Start from reference/guide-template.md in this skill directory and fill it in.

3. Section order

Every guide uses this order. Omit a section only if it genuinely does not apply; never reorder.

  1. # Losant <Tool> Tool Guide — H1 title.
  2. One or two sentences on what the tool is for and when to reach for it.
  3. Any > warning blockquotes that must be read before anything else (destructive behavior, irreversibility, one-time secrets).
  4. ## Before Calling — numbered prerequisite steps. If the tool needs an applicationId, step 1 links to the query guide rather than restating the procedure: Obtain \applicationId` by following the application-selection procedure in losant://guides/tools/losant_query.` Only the query guide spells that procedure out. Later steps walk the columns of the Resource Types table in the order the caller needs them.
  5. ## Parameters — a table, then any > blockquotes for cross-parameter rules:
    | Parameter | Required | Description |
    |---|---|---|
    
    Required is Yes, No, or Conditional. A Conditional row must say what it is conditional on, pointing at a named column of the Resource Types table ("Required only for types that name a field in the parentResourceId column below"). Descriptions reference table columns in bold rather than duplicating their contents.
  6. ## Resource Types — legend bullets, then the generated table, then the nested-type notes. One bullet per non-obvious column explaining what the value means and what — means in it. Columns are drawn from src/constants.js:
    • resource list — RESOURCE_TYPES, WRITABLE_RESOURCE_TYPES, or DELETABLE_RESOURCE_TYPES, whichever the tool accepts
    • NESTED_RESOURCES → parentResourceId column (`parentField` or —)
    • RESOURCE_TYPE_TO_GUIDE → Domain Guide column
    • RESOURCE_TYPE_TO_AUTHORING_GUIDE → Authoring Guide column (body-authoring tools only)
    • RESOURCE_TYPE_TO_QUERY_SCHEMA → Advanced Query Schema column (query-ish tools only)
    • NO_CREATE_TYPES → suppresses the Post schema link

Read the full file on GitHub · 108 lines

Files

What ships with it

1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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. 16d ago Changed 5ed7fd7ce66a
  2. 19d ago First seen · 108 lines · 70 tokens per session scan A 2125624755f5

Subscribe to this mod's changes

mcp-add-tool-guide is a skill published in the GitHub repository Losant/losant-mcp-server (2 stars, last pushed yesterday), licensed MIT. It adds 70 tokens to every session and 1,902 once invoked, about $0.0003 per session on Opus 5.5. 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 2026-10-19.

Related

mcp-add-tool-guide compared with skills from other repositories

mcp-add-tool-guide next to 6 skills from other repositories
ModTypeStarsLast pushLicenceScanTokensWorks with
mcp-add-tool-guideLosant/losant-mcp-serverSkill2yesterdayMITA70Claude Code
iot-gatewayKerberosClaw/kc_iot_gatewaySkill27mo agoMITA31Claude Code, Codex
modbusKerberosClaw/kc_modbus_mcpSkill27mo agoMITA27Claude Code, Codex
opik-diagnosecomet-ml/opik-mcpSkill225yesterdayApache-2.0A187Claude Code
ue-mcp-epic-routingdb-lyon/ue-mcpSkill480todayMITA81Claude Code, Codex
qa-knowledgegriddynamics/rosettaSkill373todayApache-2.0A32Claude Code, Codex
apitapn1byn1kt/apitapSkill13117d agoApache-2.0A0Claude Code, Codex