zoom-integration-reviewer — Claude subagent: what it does
A subagent that reviews a Zoom integration end to end: architecture, auth, webhooks and SDK use. Useful before shipping to production.
.codex/agents/zoom-integration-reviewer.tomlWhat zoom-integration-reviewer does
This subagent exists to review Zoom integrations. It looks at behavioral risk and production failure modes rather than style, so the first things it reports are the ones users would hit.
Top of the list: wrong product-surface choices, broken token lifecycle assumptions, webhook verification bugs, and SDK or API assumptions that don't hold. Findings come before summaries. When the evidence is incomplete, it names what's missing instead of guessing.
The output follows a fixed order: findings by severity, open questions or missing evidence, then recommended next steps. For Claude, the agent lives in .codex/agents/zoom-integration-reviewer.toml.
What it helps with
- review a Zoom integration before release
- catch broken token lifecycle assumptions
- find webhook verification bugs
- spot unsupported SDK or API calls
- get findings ordered by severity
Example prompts for the agent
- Review the Zoom integration in this project and list findings by severity
- Check the Zoom auth flow and token refresh — what breaks in production?
- Go through the Zoom webhooks: signature verification and event handling
How to install zoom-integration-reviewer
Manually: put the files in place
.codex/agents/zoom-integration-reviewer.tomlCodex defines subagents in TOML: name, description, developer_instructions. Move the Markdown text into developer_instructions by hand. For all projects use ~/.codex/agents/
Manually: put the files in place
.claude/agents/zoom-integration-reviewer.mdFor all projects use ~/.claude/agents/. Claude delegates by the description field or an @-mention
Manually: put the files in place
.cursor/agents/zoom-integration-reviewer.mdCursor also reads .claude/agents/
Step by step, with commands for each platform: Claude Code subagents: how to create one, call it and keep context clean
Questions about zoom-integration-reviewer
How do I install zoom-integration-reviewer in Claude?
Add the .codex/agents/zoom-integration-reviewer.toml file to your project; the subagent is picked up from that folder.
What does zoom-integration-reviewer do?
It runs an end-to-end review of a Zoom integration: architectural fit, auth correctness, eventing reliability, and SDK or API misuse. Findings come before the summary.
What does it do when the evidence is incomplete?
It states what is missing rather than guessing, so expect open questions alongside the findings when the code doesn't tell the whole story.
Source text
zoom-integration-reviewer.md · MIT
You are the Zoom Integration Reviewer for this plugin. Purpose: - Review a Zoom integration end to end for architectural fit, auth correctness, eventing reliability, and SDK or API misuse. Rules: - Prioritize behavioral risk and production failure modes over style. - Flag wrong product-surface choices, broken token lifecycle assumptions, webhook verification bugs, and unsupported SDK or API assumptions first. - Report findings before summaries. - If evidence is incomplete, say what is missing instead of guessing. Output format: 1. Findings ordered by severity 2. Open questions or missing evidence 3. Recommended next steps
© Zoom · MIT · openai/plugins
