← Files BrizzARCHIVED FILE
plugin.json
6.03 KB · Nov 16, 2026 · 18:17 UTC
{
"$schema": "https://agent-plugins.org/schemas/1.0.0/plugin.schema.json",
"name": "brizz",
"version": "1.0.0",
"description": "Product analytics for AI agents. Search your agent's user sessions, read conversation transcripts, investigate detected issues, and query usage, quality and cost metrics.",
"author": {
"name": "Brizz",
"url": "https://brizz.ai"
},
"homepage": "https://brizz.ai",
"keywords": [
"analytics",
"observability",
"ai-agents",
"llm",
"telemetry"
],
"extensions": {
"com.openai": {
"onboardingSkill": "./skills/brizz-setup/SKILL.md",
"interface": {
"displayName": "Brizz",
"shortDescription": "Analytics for your AI agents",
"longDescription": "Brizz is the product analytics layer for AI agents. It captures every conversation your deployed agents have with their users, evaluates response quality, detects failure patterns, and groups them into issues.\n\nConnect your Brizz workspace to ask about your agents in plain language:\n- Find and read user sessions and their transcripts, errors and tool calls.\n- Investigate a detected issue: what went wrong, how often, which users it hit, and the evidence behind it.\n- Query usage, quality, latency and cost metrics, broken down by service, organization, model or cohort.\n- Manage your workspace: update an issue's status or assignee, link the pull request that fixes it, and edit classifications, cohorts and dashboards.\n\nRequires a Brizz account. You choose which workspace to connect when you sign in, and every request is limited to that workspace's data.",
"developerName": "Brizz",
"category": "Developer Tools",
"capabilities": [
"Read",
"Write"
],
"websiteURL": "https://brizz.ai",
"supportURL": "https://brizz.ai/contact",
"privacyPolicyURL": "https://platform.brizz.dev/privacy-policy",
"termsOfServiceURL": "https://platform.brizz.dev/terms-of-service",
"defaultPrompt": [
"What are the top open issues in my agent this week?",
"Show me sessions from the last day where users were frustrated",
"How did my agent's cost and error rate change over the last 7 days?"
],
"brandColor": "#052129",
"brandColorDark": "#F783AC",
"composerIcon": "./assets/icon.png",
"logo": "./assets/logo.png"
},
"review": {
"test_cases": {
"positive": [
{
"description": "Top issues: list the open issues Brizz detected in the connected workspace's AI agent.",
"prompt": "What are the top open issues in my AI agent right now?",
"tools_triggered": "brizz_search_issues",
"expected_behavior": "Returns a ranked list of open issues, each with its title, category, number of affected sessions and status."
},
{
"description": "Investigate one issue: explain a detected issue's root cause using its evidence and an affected conversation.",
"prompt": "Explain the most frequent issue in my agent: what goes wrong, and show me an example conversation.",
"tools_triggered": "brizz_search_issues, brizz_get_issue, brizz_get_issue_findings, brizz_get_conversation",
"expected_behavior": "Summarizes what goes wrong in the issue, how often it happens and the evidence behind it, then quotes a short excerpt from one affected conversation."
},
{
"description": "Find sessions: filter the agent's user sessions by a quality signal over a time window.",
"prompt": "Find sessions from the last 7 days where the user was frustrated.",
"tools_triggered": "brizz_list_session_filters, brizz_search_sessions",
"expected_behavior": "Returns a list of matching sessions with date, service and a one-line summary of each, or says plainly that none matched in the window."
},
{
"description": "Metrics trend: report how usage and errors changed over time.",
"prompt": "How did my agent's session count and error rate change over the last 14 days?",
"tools_triggered": "brizz_query_metric, brizz_aggregate_sessions",
"expected_behavior": "Returns a day-by-day series for sessions and error rate and calls out the overall change between the start and end of the window."
},
{
"description": "Write action: change an issue's status and priority in the workspace.",
"prompt": "Mark the top open issue as tracked and set its priority to high.",
"tools_triggered": "brizz_search_issues, brizz_update_issue",
"expected_behavior": "Confirms the issue's status and priority were changed; the change is visible on the Issues page of the Brizz dashboard."
}
],
"negative": [
{
"description": "Unrelated request (weather). Brizz tools should not be called; the model answers without the plugin or says it cannot help.",
"prompt": "What's the weather in Paris tomorrow?"
},
{
"description": "General coding question with no Brizz data involved. Brizz tools should not be called; the model answers on its own.",
"prompt": "Write a Python function that reverses a linked list."
},
{
"description": "Request for another company's data. Every token is scoped to the one workspace chosen at sign-in, so no other tenant's data can be returned; the model explains it can only see the connected workspace.",
"prompt": "Show me the conversations of another company that uses Brizz."
}
]
}
},
"publication": {
"release_notes": "Initial release: search sessions and transcripts, investigate detected issues, query agent metrics, and manage issues, classifications, cohorts and dashboards in your Brizz workspace."
},
"apps": "./.app.json"
}
}
}
SHA-256: bd39415bb114372e99fd62367c9055962305734303d8800628e85f547c2f75dc