skills.rest
iamtouchskyer
iamtouchskyerCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

opc

Coordinate a multi-agent digraph review pipeline with gate verdicts.

189|41|Updated Mar 27, 2026
Search Tags:#workflow#multi-agent#code-review#gate#ai-review#handshake#opc
One-click install
npx skills add https://github.com/iamtouchskyer/opc --skill opc
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: opc
Source: https://github.com/iamtouchskyer/opc/tree/main
Command: npx skills add https://github.com/iamtouchskyer/opc --skill opc

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

Multi-role evaluation for code and product ideas is time-consuming and error-prone; OPC coordinates a digraph-based, multi-agent workflow that orchestrates specialized agents to build, review, and evaluate tasks with independent verdicts and traceable provenance.

Core Features & Use Cases

  • Digraph engine with typed nodes (build, review, gate, loop) and mechanical gate verdicts to enforce objective quality gates
  • Autonomous loop (opc loop) that decomposes features, schedules durable cron, and runs unattended with guardrails
  • External validator integrations (pre-commit hooks, tests, E2E) to supervise the agent workflow
  • Protocols and handshakes that produce machine-readable state in .harness for downstream tooling

Quick Start

Use the OPC skill to orchestrate multi-agent evaluation on your codebase by initiating an appropriate flow and letting OPC dispatch the required agents automatically.

Frequently Asked Questions about opc

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

FAQPage Schema
How do I run a multi-agent code review pipeline?▼

Run a multi-agent code review pipeline by initiating a flow like build-verify or full-stack, letting OPC dispatch 11 specialist agents automatically to build, review, and evaluate code with deterministic gate verdicts.

What are deterministic gate verdicts in an AI review workflow?▼

Deterministic gate verdicts are mechanical quality gates enforced by a digraph engine with typed nodes, ensuring objective evaluation and traceable provenance across specialized agent handshakes.

Can I automate feature decomposition and unattended code evaluation?▼

Yes, you can automate feature decomposition and unattended code evaluation using the autonomous loop, which schedules durable cron tasks and runs guardrailed workflows without manual intervention.

Does the multi-agent review workflow integrate with pre-commit hooks and E2E tests?▼

Yes, the multi-agent review workflow integrates external validators like pre-commit hooks, tests, and E2E suites to supervise the agent workflow and enforce quality gates.

What is the best way to orchestrate independent agent verdicts for code evaluation?▼

The best way to orchestrate independent agent verdicts is using a digraph-based multi-agent workflow that coordinates 11 specialist agents to build, review, and evaluate tasks with traceable provenance.

How do downstream tools validate progress from a multi-agent code review?▼

Downstream tools validate progress by reading machine-readable flow-state and handshake artifacts output under the .harness directory, which record deterministic gate verdicts and pipeline state.

Related Skills

Explore More Software Engineering Skills→
TeiNamTeiNam
12

security-scan

Scan agent configurations for security vulnerabilities and misconfigurations.

Community
Intermediate
dodygdodyg

dotnet-benchmarkdotnet

Automate .NET microbenchmark measurement and analysis with BenchmarkDotNet.

Community
Intermediate
liauw-medialiauw-media
3

pre-commit-hooks

Configure pre-commit hooks for JavaScript/TypeScript, Python, and PHP projects.

Community
Advanced
RangeyRoverRangeyRover
3

speckit.checker

Run static analysis across multiple languages and generate prioritized issue reports.

Community
Intermediate