relay-80-100-workflow

Validate agent-relay workflows end-to-end with PGlite and mock sandboxes.

28|4|Updated Dec 20, 2025
One-click install
npx skills add https://github.com/AgentWorkforce/trajectories --skill relay-80-100-workflow
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: relay-80-100-workflow
Source: https://github.com/AgentWorkforce/trajectories/tree/main/.agents/skills/relay-80-100-workflow
Command: npx skills add https://github.com/AgentWorkforce/trajectories --skill relay-80-100-workflow

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Writes agent-relay workflows that must fully validate features end-to-end before merging, ensuring the feature works and has passing tests rather than only compiling.

Core Features & Use Cases

  • End-to-end workflow templates that integrate with deterministic tests, in-memory Postgres (PGlite), mock sandboxes, and a complete lifecycle from implementation to commit.
  • Verification gates after every edit to ensure changes actually land and don't regress.
  • The three-step test-fix-rerun pattern and commit gating to maintain production-readiness.

Quick Start

Implement an 80-to-100 workflow by running deterministic tests, fixing failures, and gating commits on green results.

Frequently Asked Questions about relay-80-100-workflow

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

FAQPage Schema
How do I ensure end-to-end validation passes before merging agent-relay workflows?

End-to-end validation for agent-relay workflows is ensured by applying an 80-to-100 pattern, which runs deterministic tests using in-memory Postgres (PGlite) and mock sandboxes to verify features before merging.

What is the 80-to-100 pattern for local workflow testing?

The 80-to-100 pattern is a local testing strategy that takes features from an 80% complete state to fully validated by running deterministic tests, fixing failures, and gating git commits until all results are green.

How do I gate git commits on passing tests in an automated workflow?

Git commit gating is implemented by integrating a test-fix-rerun loop that blocks commits until all deterministic tests pass, ensuring only verified and production-ready changes are merged.

Can I use in-memory Postgres and mock sandboxes for deterministic testing?

In-memory Postgres (PGlite) and mock sandboxes are used to provide isolated, deterministic testing environments that validate agent-relay workflows without requiring external database dependencies.

Why do I need verification gates after every edit in an agent-relay workflow?

Verification gates after every edit are needed to confirm that code changes actually land correctly and do not introduce regressions, maintaining continuous production-readiness throughout the workflow lifecycle.