iterative-retrieval

Coordinates sub-agent task delegation with a 3-cycle retry protocol and validation checklist.

2|Updated Jul 24, 2026
One-click install
npx skills add https://github.com/elbruno/ElBruno.MagenticUI --skill iterative-retrieval-elbruno
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: iterative-retrieval
Source: https://github.com/elbruno/ElBruno.MagenticUI/tree/main/.squad/templates/skills/iterative-retrieval
Command: npx skills add https://github.com/elbruno/ElBruno.MagenticUI --skill iterative-retrieval-elbruno

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? When coordinators spawn sub-agents for scoped work, handoffs often lack context, retry cycles multiply without limit, and outputs get accepted without verification. This Skill enforces a structured delegation protocol that caps retries at 3 cycles, mandates WHY context in every spawn prompt, and requires coordinator validation before closing work. ## Core Features & Use Cases - Structured Spawn Prompts: A four-section template (Task, WHY, Success Criteria, Escalation Path) ensures every sub-agent receives complete context and explicit acceptance criteria. - 3-Cycle Retry Protocol: Each retry carries forward a delta summary of what was tried and what remains, with mandatory escalation after cycle 3 instead of endless loops. - Coordinator Validation & Issue Dedup: A pre-acceptance checklist plus a mandatory gh issue list search prevents duplicate GitHub issues and unvalidated output. - Use Case: A coordinator agent delegates a documentation update to a sub-agent, validates the PR against the success criteria checklist, spawns one targeted retry for a missing section, then closes the issue after verification. ## Quick Start Use the iterative-retrieval protocol to spawn a sub-agent for this task with a full WHY-context prompt and validate its output against the success criteria before accepting it.

Frequently Asked Questions about iterative-retrieval

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

FAQPage Schema
How do I structure a prompt when spawning a sub-agent?

Use the four-section spawn template: Task (concrete and bounded), WHY this matters (motivation and context), Success criteria (explicit checkboxes), and Escalation path (what to do when stuck). Vague prompts without WHY context are listed as an anti-pattern.

How many retry cycles should a sub-agent get before escalating?

The protocol caps retries at 3 cycles. Each subsequent cycle must include a delta summary of what was tried and what remains. After cycle 3, escalate by writing a summary to the decisions inbox and labeling the issue status:needs-decision.

How do I avoid creating duplicate GitHub issues with agents?

Run gh issue list --search with 2-3 keywords from the planned issue title before creating any issue. If an open issue already covers the problem, comment on it instead of creating a new one.

When should I escalate instead of retrying a sub-agent task?

Escalate when requirements are fundamentally unclear, 3 cycles complete without convergence, results conflict across cycles, or the task needs elevated permissions. Retry only when output is structurally correct with identifiable remaining deltas.

What should a coordinator check before accepting sub-agent output?

Verify all success criteria are met, the PR exists and matches the issue, no regressions or new TODO/FIXME markers appear, no task parts were silently skipped, and any reported uncertainty was resolved or escalated.