f1-test-drive

Validates the Cyrus agent pipeline end-to-end across issue tracking, EdgeWorker sessions, and activity rendering.

794|163|Updated Apr 19, 2025
One-click install
npx skills add https://github.com/ceedaragents/cyrus --skill f1-test-drive
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: f1-test-drive
Source: https://github.com/ceedaragents/cyrus/tree/main/skills/f1-test-drive
Command: npx skills add https://github.com/ceedaragents/cyrus --skill f1-test-drive

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Verifying that the Cyrus agent system works end-to-end requires manually checking issue creation, agent session execution, and activity output across multiple components, which is slow and error-prone without a structured test protocol.

Core Features & Use Cases

  • End-to-End Pipeline Validation: Runs a phased test drive covering issue-tracker behavior, EdgeWorker worktree/session execution, and activity rendering quality.
  • Slack Chat Session Verification: Optionally validates the Slack to ChatSessionHandler to ClaudeRunner path using a synthetic app_mention dispatch endpoint, including shared auto-memory and per-thread workspace isolation checks.
  • Structured Reporting: Produces standardized test drive reports under apps/f1/test-drives/ with checklists, session logs, and retrospectives.
  • Use Case: Before merging a major runner or harness change, run an F1 test drive to confirm the server starts, issues are created, sessions produce coherent activities, and everything stops cleanly.

Quick Start

Run an F1 test drive to validate the full Cyrus pipeline from issue creation through agent session activity rendering and write the report.

Frequently Asked Questions about f1-test-drive

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run an end-to-end test of the Cyrus agent system?

Start the F1 server on port 3600 with a fresh test repository, create a test issue, start an agent session, and monitor activities with the f1 view-session command. Finish by stopping the session and writing a report under apps/f1/test-drives/.

How do I test the Slack chat session path without a real Slack workspace?

Use the F1 test-only /cli/dispatch-chat endpoint via f1 start-chat-session, which injects a synthetic app_mention event without Slack signature verification. The SlackChatAdapter no-ops Slack API calls when no bot token is configured.

What are the pass and fail criteria for an F1 test drive?

A drive passes when the server starts, an issue is created, a session starts with visible coherent activities, and the session stops cleanly without unhandled errors. It fails on startup errors, failed issue creation, missing activities, malformed payloads, or unhandled exceptions.

Can Claude, Codex, or other agent harnesses use this test drive skill?

Yes, the skill is intentionally harness-agnostic. Claude subagents, Codex, and OpenCode workflows can all reference the same skill content, with harness-specific adapters acting as thin wrappers.

When should I run an F1 test drive before merging changes?

Run at least one F1 end-to-end validation before merging any major runner or harness change. Use fresh test repositories per drive and preserve failed state when debugging issues.