create-new-mcp

create-new-mcp is a skill for Claude Code, Codex from XAIHT/Tlamatini. It costs 147 tokens per session (1,969 once invoked), scanned A, original, MIT.

A reference procedure for adding a new tool, MCP data provider or both to the Tlamatini application. It explains how to classify the request and connect the implementation to the application's tools, database and configuration.

In plain words
What is it for?
Use it when adding a tool function, creating an MCP context provider, registering database records, extending the factory and setting the required application controls.
Why use it?
It reduces the risk of registering an integration in the wrong place or missing required backend setup. It also clarifies when a request needs an on-demand tool, preloaded context, or both.

Skill for Claude CodeCodex

Which agent this was written for is unclear — built for openclaw. Also seen: built for openclaw.

Good fit Use it when adding a tool function, creating an MCP context provider, registering database records, extending the factory and setting the required application controls.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/xaiht/tlamatini/create_new_mcp
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.

Any agent
npx skills add XAIHT/Tlamatini --skill create_new_mcp
Clone the repo
git clone --depth 1 https://github.com/XAIHT/Tlamatini

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 create-new-mcp

README.md
[![agentmods](https://agentmods.dev/badge/skills/xaiht/tlamatini/create_new_mcp/github.svg)](https://agentmods.dev/skills/xaiht/tlamatini/create_new_mcp)
Your own site
<a href="https://agentmods.dev/skills/xaiht/tlamatini/create_new_mcp"><img src="https://agentmods.dev/badge/skills/xaiht/tlamatini/create_new_mcp/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 create-new-mcp

Your own site · 80×15
<a href="https://agentmods.dev/skills/xaiht/tlamatini/create_new_mcp"><img src="https://agentmods.dev/badge/skills/xaiht/tlamatini/create_new_mcp.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 147 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,969 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. Third-party audits
  • NVIDIA SkillSpector warn 7 Sept 2026
SkillSpector: 1 finding, up to medium

These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →

  • medium MCP Rug Pull · line 107
    uvx/uv tool run commands without ==version create a rug-pull risk.
    Fix: Pin the version: uvx package-name==1.2.3
How audits are shown
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 $0.00147 $0.01969
Opus 5 $0.00073 $0.00984
Sonnet 5 $0.00029 $0.00394
Haiku 4.5 $0.00015 $0.00197

Measured 9d ago against content hash a40449f7987f, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

create-new-mcp 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 9d 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.

Tlamatini/agent/skills_pkg/create_new_mcp/SKILL.md · 124 lines

How it starts

The opening of the file, as written. The whole thing — 124 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Create New MCP / Tool — canonical reference

The full, ground-truth procedure for adding a tool, an MCP context provider, or both lives at:

Tlamatini/.mcps/create_new_mcp.md

First decision — classify the request before writing anything

Type The request looks like... Path to follow
Tool only "let the LLM do X on demand" — run a command, start an agent, unzip, decompile, summarize. Tool-only workflow (below).
MCP-backed context provider only "every chat turn, prefetch X and inject it as context" — system metrics, files search, repo inventory. MCP context-provider workflow (below).
Both Pre-fetched context and a separate action tool. Rare; most are one or the other. Do both halves.

Tool-only workflow (most common)

  1. Implement an @tool function in agent/tools.py (sync, returns strings).
  2. Resolve any bundled paths for both source and frozen modes (os.path.dirname(sys.executable) vs os.path.dirname(os.path.abspath(__file__))).
  3. Register the tool in get_mcp_tools() under a global_state gate.
  4. Seed a Tool row via a new migration so the UI toggle row exists.
  5. Frontend usually needs NO changes — the tool checkbox UI is dynamic.
  6. The tool is only usable when the unified-agent / Multi-Turn chain is selected.

MCP context-provider workflow

  1. Create mcp_<name>_server.py + mcp_<name>_client.py.
  2. Create a sidecar chain chain_<name>_lcel.py.
  3. Wire startup in apps.py + management/commands/startserver.py.
  4. Extend rag/factory.py: import the chain, add the sync wrapper, map the status key, patch invoke() to inject the new payload field.
  5. Choose a payload field and update all main chains (basic / history-aware / unified).
  6. Seed an Mcp row and update the frontend MCP checkboxes (they are hardcoded, NOT dynamic — unlike Tool checkboxes).
  7. Verify persistence and reconnect behavior end-to-end.

Read the full file on GitHub · 124 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. 9d ago First seen · 124 lines · 147 tokens per session scan A a40449f7987f

Subscribe to this mod's changes

create-new-mcp is a skill published in the GitHub repository XAIHT/Tlamatini (16 stars, last pushed today), licensed MIT. It adds 147 tokens to every session and 1,969 once invoked, about $0.0007 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 2026-08-30.