make-plugin

make-plugin is a skill for Claude Code, Codex from smixs/iva-agent. It costs 62 tokens per session (1,006 once invoked), scanned A, original, MIT.

A guide for creating an add-on in the Agent Plugins format, usually from a skill folder, scripts, and a configuration file. It explains when a persistent MCP server or custom code is needed instead of a simple script.

In plain words
What is it for?
Use it when building a new plugin, teaching the agent a repeatable task, connecting an external service, or learning how plugins are written.
Why use it?
It provides a defined structure for turning a requested capability or connection into an installable add-on. It also requires scripts to be run before installation and separates changes that need the owner's approval.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one.

Good fit Use it when building a new plugin, teaching the agent a repeatable task, connecting an external service, or learning how plugins are written.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/smixs/iva-agent/make-plugin
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 smixs/iva-agent --skill make-plugin
Clone the repo
git clone --depth 1 https://github.com/smixs/iva-agent

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 make-plugin

README.md
[![agentmods](https://agentmods.dev/badge/skills/smixs/iva-agent/make-plugin/github.svg)](https://agentmods.dev/skills/smixs/iva-agent/make-plugin)
Your own site
<a href="https://agentmods.dev/skills/smixs/iva-agent/make-plugin"><img src="https://agentmods.dev/badge/skills/smixs/iva-agent/make-plugin/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 make-plugin

Your own site · 80×15
<a href="https://agentmods.dev/skills/smixs/iva-agent/make-plugin"><img src="https://agentmods.dev/badge/skills/smixs/iva-agent/make-plugin.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 62 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,006 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 $0.00062 $0.01006
Opus 5.5 $0.00025 $0.00402
Sonnet 5.5 $0.00012 $0.00201
Haiku 4.5 $0.00006 $0.00101

Measured 7d ago against content hash cf2119e4cd2e, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

make-plugin 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 7d 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.

agent/skills/make-plugin/SKILL.md · 61 lines

How it starts

The opening of the file, as written. The whole thing — 61 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Make a plugin

A new capability is a plugin: a folder in the Agent Plugins format (ADR-0008, ADR-0009). The full format is in your own documentation, docs/plugins.md in the working directory of the running version ("Write your own"); read it before the first plugin of a turn.

1. Build the draft

  • Folder: data/custom/plugin-drafts/<name>/. <name> is lowercase letters, digits and hyphens, starting and ending with a letter or a digit.
  • plugin.json with $schema (https://agent-plugins.org/schemas/1.0.0/plugin.schema.json), name (the folder name), version, description.
  • The capability is a skill: skills/<skill>/SKILL.md with name and a trigger description ("Use when…"). Its tools are scripts beside it (skills/<skill>/scripts/) with a README.md that says what each one takes and prints. Run every script once before you install.
  • MCP (mcp.json) or code under sh.iva/ only when a script cannot do the job: a server that must stay up, a tool the model calls on its own. Each of them makes the install wait for the owner's tap.
  • A script that just runs and prints is not a service: keep it in the skill's scripts/ and install without a tap. sh.iva/ is only an Extension (sh.iva/package.json, TypeScript) or a service — a process that stays up: sh.iva/services/<svc>/service.json with command and port (1024–65535), listening on 127.0.0.1:$IVA_SERVICE_PORT. propose refuses a draft with a part the reader would drop and names the line.
  • No symlinks, no secrets in files: keys go to .env or data/plugin-data/<name>/, named in the README.

2. Install

In an Insight turn you only build and run the draft; installing waits for the owner's tap (skill insight).

  • Skills and scripts only (no mcp.json, no sh.iva/): iva plugin add ./data/custom/plugin-drafts/<name> from bash. It works from the next turn: no build, no restart. Say so.
  • With mcp.json or sh.iva/: iva plugin propose ./data/custom/plugin-drafts/<name>. The code sends the owner a message with what the plugin will run and an Install button. End the turn with "sent the proposal, waiting for the tap" in the owner's language ("отправила предложение, жду тапа"). The result arrives later as a message from the code; do not promise it and do not check on it.
  • Never install such a plugin any other way: iva plugin add refuses it outside the owner's terminal, and iva plugin trust|enable|update|sync|install-proposal are blocked in bash. An update of an installed plugin is iva plugin update <name> in the owner's terminal: tell the owner.
  • A refusal from propose or add is plain text: fix the draft (the diagnostic names the file) and run it again.

Read the full file on GitHub · 61 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. 7d ago Changed · +2 lines cf2119e4cd2e
  2. 10d ago First seen · 59 lines · 62 tokens per session scan A dd0a72970656

Subscribe to this mod's changes

make-plugin is a skill published in the GitHub repository smixs/iva-agent (240 stars, last pushed today), licensed MIT. It adds 62 tokens to every session and 1,006 once invoked, about $0.0002 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 2026-10-06.