skills.rest
Paulwhoisaghostnet
PaulwhoisaghostnetCOMMUNITY
Explore all skills by this author →
GitHub

subagent_skill_harvest_loop

Orchestrate subagent rounds to generate prototypes, harvest evidence, and patch skills.

2|Updated Apr 13, 2026
Search Tags:#feedback#subagent#prototype#patch#workspaces#harvest
One-click install
npx skills add https://github.com/Paulwhoisaghostnet/kiln --skill subagent-skill-harvest-loop
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: subagent_skill_harvest_loop
Source: https://github.com/Paulwhoisaghostnet/kiln/tree/main/skllz/skills/subagent_skill_harvest_loop
Command: npx skills add https://github.com/Paulwhoisaghostnet/kiln --skill subagent-skill-harvest-loop

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Delegated subagent rounds enable rapid prototype generation, evidence collection, feedback synthesis, and skill patches while keeping workspaces clean.

Core Features & Use Cases

  • Delegated rounds coordinate multiple subagents to explore task families and produce artifacts.
  • Evidence harvesting collects prototypes, feedback, and documentation to drive skill improvements.
  • Skill patching applies validated learnings to references and ensures parity across skills.
  • Works with disjoint ownership, artifact schemas, and structured cleanup to enable repeatable cycles.

Quick Start

Define the round objective, spawn isolated per-agent workspaces, run subagents, collect prototypes and feedback, patch the skills, and clean up temporary workspaces.

Frequently Asked Questions about subagent_skill_harvest_loop

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

FAQPage Schema
How do I coordinate multiple subagents to collect prototypes and feedback?▼

To coordinate multiple subagents for prototype and feedback collection, you orchestrate delegated subagent rounds that spawn isolated workspaces, execute defined objectives, and harvest artifacts to drive iterative skill improvements.

How do I patch skills with new knowledge from multiple agents?▼

To patch skills with new knowledge, you apply validated learnings harvested from subagent rounds to references, which ensures parity across skills while maintaining disjoint ownership and structured workspace cleanup.

Can I use disjoint ownership when running subagent rounds for skill improvement?▼

Yes, disjoint ownership is explicitly supported when running subagent rounds for skill improvement. Multiple agents can operate under disjoint ownership on defined objectives and artifact schemas to enable repeatable improvement cycles.

What is the best way to manage isolated workspaces during iterative prototype generation?▼

The best way to manage isolated workspaces during iterative prototype generation is to spawn per-agent workspaces for each round, run the subagents to explore task families, and apply structured cleanup afterward to keep environments clean.

Do I need defined artifact schemas to harvest feedback from delegated subagent rounds?▼

Yes, defined artifact schemas are required to harvest feedback effectively. They ensure that prototypes and feedback collected from isolated workspaces can be validated and properly applied to patch skills with new knowledge.

Related Skills

Explore More Software Engineering Skills→
badaniyabadaniya
1

subagent-driven-development

Dispatch a fresh subagent per task and run two-stage reviews.

Community
Intermediate
ArtyMcLabinArtyMcLabin
8

retrospective

Generate a structured retrospective template from session notes after mistakes.

Community
Basic
irahardiantoirahardianto
150

parallel-dispatch-decomposition

Decompose software tasks into MECE parallelizable sub-tasks with scope cards.

Community
Advanced
richkuorichkuo
45

fix-pr-review-loop

Automate pull request review feedback resolution via GitHub CLI.

Community
Advanced