connector-generator

connector-generator is an agent for coding agents from fivetran/connector_sdk_tools. It costs 22 tokens per session (3,837 once invoked), scanned A, original, MIT.

A code-generation agent that adapts an existing Fivetran Connector SDK project from a completed connector specification.

In plain words
What is it for?
Use it after requirements gathering to adapt scaffolded connector files for authentication, endpoints, pagination, schemas, and related behavior.
Why use it?
It turns already-validated requirements into targeted changes while preserving the generated project’s working structure.

Agent

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.

agentmods
npx agentmods add agents/fivetran/connector_sdk_tools/connector-generator
Clone the repo
git clone --depth 1 https://github.com/fivetran/connector_sdk_tools

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 connector-generator

README.md
[![agentmods](https://agentmods.dev/badge/agents/fivetran/connector_sdk_tools/connector-generator.svg)](https://agentmods.dev/agents/fivetran/connector_sdk_tools/connector-generator)
Your own site
<a href="https://agentmods.dev/agents/fivetran/connector_sdk_tools/connector-generator"><img src="https://agentmods.dev/badge/agents/fivetran/connector_sdk_tools/connector-generator.svg" alt="Measured on agentmods" height="20"></a>
Per session 22 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 3,837 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00022 $0.03837
Opus 5 $0.00011 $0.01919
Sonnet 5 $0.00004 $0.00767
Haiku 4.5 $0.00002 $0.00384

Measured 4d ago against content hash 9cdf75a2a8fe, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

connector-generator 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 4d 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/connector-generator.md · 348 lines

How it starts

The opening of the file, as written. The whole thing — 348 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Fivetran Connector Code Generation

FIRST: Read sdk-reference.md from the plugin directory to load SDK rules, patterns, and example URLs.

Where to look: patterns & examples → connector_sdk (exhaustive). Community connectors → community_connectors.

YOUR ROLE: ADAPT THE SCAFFOLDED CONNECTOR

The project has already been scaffolded by fivetran initconnector.py, configuration.json, and a dependency file already exist on disk. For a community-connector template, connector.py already contains a real, working implementation; for the default template, it contains a complete runnable starter (validate_configuration(), schema(), update(), docstrings, the __main__ block).

You also receive a COMPLETE specification from the validation phase. The validator has already researched the API and identified auth/endpoints/pagination.

Your job is to ADAPT the existing scaffolded files to that specification using Edit — not to rewrite them from scratch. Do NOT re-research the API. Read the scaffolded files first, then make targeted edits. Preserve the template's structure: validate_configuration() called at the top of update(), function docstrings, the global connector = Connector(...), and the __main__ block. Use Write only for a file the scaffold did not create.

Mandatory Example Analysis

Before writing code, use WebFetch to study 2-4 relevant SDK examples (see example URLs in sdk-reference.md):

  1. Always fetch the hello world example for basic structure
  2. Fetch the authentication example matching the API's auth method
  3. Fetch pagination example if needed
  4. Document what you learned before coding:
    Examples studied:
    - [URL]: [key pattern learned]
    Implementation approach:
    - Authentication: [method] following [example]
    - Pagination: [type] based on [example]
    

Read the full file on GitHub · 348 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. 4d ago First seen · 348 lines · 22 tokens per session scan A 9cdf75a2a8fe

Subscribe to this mod's changes

connector-generator is an agent published in the GitHub repository fivetran/connector_sdk_tools (87 stars, last pushed 1mo ago), licensed MIT. It adds 22 tokens to every session and 3,837 once invoked, about $0.0001 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.