skills.rest
toqoz
toqozCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

session-harness-runner

Execute and verify session harness packages with iteration control and convergence detection.

Updated Oct 15, 2021
Search Tags:#automation#workflow#testing#verification#agent#harness
One-click install
npx skills add https://github.com/toqoz/config --skill session-harness-runner
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: session-harness-runner
Source: https://github.com/toqoz/config/tree/main/home/agents/skills/session-harness-runner
Command: npx skills add https://github.com/toqoz/config --skill session-harness-runner

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This skill resolves the complexity of managing agent-based task execution by providing a standardized, self-contained environment for running, verifying, and iterating on session harnesses.

Core Features & Use Cases

  • Unified Execution: Provides a single entry point to trigger agent invocation, verification, and convergence detection.
  • Configurable Workflows: Supports multiple execution modes including one-shot, verify-only, and bounded-loop iterations.
  • Use Case: Use this skill to run a specific task harness when you need to verify agent performance or generate a retrospective report on a completed session.

Quick Start

Execute the session harness runner by running the bash script located in the session harness directory for your specific task slug.

Frequently Asked Questions about session-harness-runner

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

FAQPage Schema
How do I run and verify agent task workflows in a bash environment?▼

Run and verify agent task workflows by executing the bash-compatible harness run script located in your task-specific session directory. This triggers agent invocation, convergence detection, and retrospective generation.

What is a session harness for automated agent testing?▼

A session harness is a self-contained package standardizing agent-based task execution, verification, and iteration. It resolves workflow complexity by providing a unified environment for running and validating agent performance.

Can I configure workflow iteration modes for bounded loops or one-shot verification?▼

Workflow iteration supports configurable execution modes including one-shot, verify-only, and bounded-loop iterations. These modes manage iteration control and convergence detection for agent-based tasks.

Do I need a specific environment to execute self-contained session harnesses?▼

Executing self-contained session harnesses requires a bash-compatible environment to run the harness script and process task-specific configuration files for agent invocation and workflow management.

How do I generate a retrospective report on a completed agent session?▼

Generate a retrospective report on a completed agent session by triggering the session harness runner. It manages task workflow iteration and automatically produces retrospective outputs upon session completion.

What's the best way to automate convergence detection during agent task execution?▼

Automate convergence detection during agent task execution by utilizing the built-in iteration control within the session harness runner. It monitors self-contained workflows and detects task convergence automatically.

Related Skills

Explore More Software Engineering Skills→
luisgustavooliveiraluisgustavooliveira

setup-gbrain

Automate gbrain setup with CLI installation, brain initialization, MCP registration, and trust policy configuration.

Community
Advanced
muratcankoylanmuratcankoylan
17.7k

reasoning-trace-optimizer

Analyze interleaved thinking traces to detect failure patterns and optimize prompts.

Community
Advanced
OceanEyeFFOceanEyeFF
29

generic-worker-skill

Execute bounded Harness tasks and output a structured execution report.

Community
Intermediate
AGMO-IncAGMO-Inc
4

verification

Automate task completion verification by collecting fresh evidence after each TODO.

Official
Advanced