bulk-evaluate

Split large tasks into sub-tickets for independent agent analysis.

Updated Mar 3, 2026
One-click install
npx skills add https://github.com/tarrragon/ccsession --skill bulk-evaluate
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: bulk-evaluate
Source: https://github.com/tarrragon/ccsession/tree/main/.claude/skills/bulk-evaluate
Command: npx skills add https://github.com/tarrragon/ccsession --skill bulk-evaluate

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Context-limited main thread cannot handle large-scale analysis within a single task. This tool splits the work into N sub-tickets, enabling each to be independently analyzed by an Agent while the main thread only collects summary results.

Core Features & Use Cases

  • Task decomposition: break large workflows into discrete, trackable sub-tickets.
  • Context unloading: offload heavy data reading and analysis to Agents, preventing main-thread context explosion.
  • Consistent conclusions: Each sub-ticket records a clear conclusion and findings; the parent ticket aggregates statistics.

Quick Start

Describe how to split a large task into N sub-tickets and delegate analysis to independent agents.

Frequently Asked Questions about bulk-evaluate

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

FAQPage Schema
How do I split large tasks into sub-tickets to unload agent context?▼

Task splitting unloads context by breaking large workflows into N independent sub-tickets, enabling agents to analyze discrete data chunks independently while the main thread collects aggregated summary statistics.

What is context unloading for batch file evaluations?▼

Context unloading offloads heavy data reading and analysis from the main thread to independent agents, preventing context explosion during batch file evaluations and large-scale reviews by recording concise conclusions in each sub-ticket.

How do I delegate analysis to agents for large-scale reviews?▼

Delegate large-scale reviews by decomposing the workload into trackable sub-tickets with standardized acceptance criteria, dispatching each to an agent for independent analysis, and aggregating the recorded conclusions in the parent ticket.

Can I use task splitting for batch processing with a context-limited main thread?▼

Yes, task splitting is designed for context-limited main threads, enabling batch processing by enforcing per-ticket analysis and concise conclusion recording, allowing the main thread to rely solely on aggregated statistics.

When should I not split workflows into independent sub-tickets?▼

Avoid splitting workflows into sub-tickets when tasks require tightly coupled, interdependent analysis across all data simultaneously, as enforcing per-ticket isolation prevents cross-context reasoning needed for holistic conclusions.

Does bulk-evaluate enforce standardized acceptance criteria for agent dispatch?▼

Yes, bulk-evaluate enforces standardized acceptance criteria for agent dispatch, ensuring each sub-ticket records a clear, consistent conclusion and findings for the parent ticket to aggregate accurately.