Orchestrate
Orchestrate three or more coding agents in parallel—research swarms, feature fan-out, or build-review-fix waves—without losing context or serializing independent work.
Overview
Orchestrate is a journey-wide agent skill that maps multi-agent dispatch strategies to work shape—usable whenever a solo builder needs to fan out 3+ agents without losing coherence.
Install
npx skills add https://github.com/hyperb1iss/hyperskills --skill orchestrateWhat is this skill?
- Strategy picker for research swarms, epic parallel builds, sequential pipelines, partitioned transforms, audits, and gre
- Patterns distilled from 597+ production agent dispatches with prompt structure mapped to agent type
- Partitions work by independence so parallelism is real, not cosmetic duplicate chats
- Build-review-fix pipelines with review overhead that can shrink as trust increases
- Triggers on swarm, fan-out, wave dispatch, multi-agent, and parallel feature build language
- Reference corpus: 597+ real agent dispatches
- Activation threshold: tasks requiring 3+ agents
- Multiple named strategies including Research Swarm and Epic Parallel Build
Adoption & trust: 602 installs on skills.sh; 13 GitHub stars; 2/3 security scanners passed (skills.sh audits).
What problem does it solve?
You have enough agent capacity to parallelize work but dispatch devolves into duplicated effort, missing context, or serial tasks disguised as “multi-agent.”
Who is it for?
Builders running Claude Code, Cursor, or Codex on large refactors, research-heavy specs, or multi-surface feature epics that need explicit orchestration.
Skip if: Single-file fixes, one clear sequential task, or sessions where coordination overhead would exceed doing the work yourself once.
When should I use this skill?
Orchestrating multi-agent work at scale—research swarms, parallel feature builds, wave dispatch, build-review-fix pipelines—or any task needing 3+ agents; mentions of swarm, fan-out, orchestrate, or parallel agents.
What do I get? / Deliverables
You pick an orchestration pattern (swarm, parallel epic, pipeline, or hybrid), partition independent slices, and run review-fix loops with proportionate oversight.
- Chosen orchestration strategy aligned to work shape
- Per-agent prompts with injected shared context and review gates
Recommended Skills
Journey fit
Useful at every journey phase - explore requirements and options before committing to a direction.
Where it fits
Launch a research swarm where each agent owns a competitor surface and merges into one decision brief.
Run an epic parallel build with one agent per feature slice and a review agent on a shared checklist.
Pipeline implementation agents with a fix wave that only revisits partitions that failed review gates.
Partition recurring chores across agents while keeping a single dependency-ordered integration step.
How it compares
Use instead of ad-hoc “open five chats” parallelism without strategy selection, context injection, or dependency-aware review.
Common Questions / FAQ
Who is orchestrate for?
Solo and small-team builders who routinely run multiple coding agents and need production-tested dispatch patterns rather than improvised fan-out.
When should I use orchestrate?
During Idea research swarms and competitor deep-dives; Validate when parallel prototypes need isolated agents; Build for epic parallel features and audits; Ship for build-review-fix waves; Operate when partitioned maintenance tasks can run concurrently.
Is orchestrate safe to install?
Orchestration implies agents may use shell, git, and network depending on your setup—review the Security Audits panel on this Prism page and scope permissions per dispatch.
SKILL.md
READMESKILL.md - Orchestrate
# Multi-Agent Orchestration Meta-orchestration patterns mined from 597+ real agent dispatches across production codebases. The skill maps orchestration strategy to work shape, prompt structure to agent type, and background/foreground to dependency graph. **Core principle:** Match the strategy to the work, partition agents by independence, inject enough context that parallelism is real, and let review overhead adapt as trust earns itself. The strategies below are reference patterns. Pick the one that fits, blend two when the work is mixed, invent your own when the patterns don't match. ## Strategy Selection ```dot digraph strategy_selection { rankdir=TB; "What type of work?" [shape=diamond]; "Research / knowledge gathering" [shape=box]; "Independent feature builds" [shape=box]; "Sequential dependent tasks" [shape=box]; "Same transformation across partitions" [shape=box]; "Codebase audit / assessment" [shape=box]; "Greenfield project kickoff" [shape=box]; "Research Swarm" [shape=box style=filled fillcolor=lightyellow]; "Epic Parallel Build" [shape=box style=filled fillcolor=lightyellow]; "Sequential Pipeline" [shape=box style=filled fillcolor=lightyellow]; "Parallel Sweep" [shape=box style=filled fillcolor=lightyellow]; "Multi-Dimensional Audit" [shape=box style=filled fillcolor=lightyellow]; "Full Lifecycle" [shape=box style=filled fillcolor=lightyellow]; "What type of work?" -> "Research / knowledge gathering"; "What type of work?" -> "Independent feature builds"; "What type of work?" -> "Sequential dependent tasks"; "What type of work?" -> "Same transformation across partitions"; "What type of work?" -> "Codebase audit / assessment"; "What type of work?" -> "Greenfield project kickoff"; "Research / knowledge gathering" -> "Research Swarm"; "Independent feature builds" -> "Epic Parallel Build"; "Sequential dependent tasks" -> "Sequential Pipeline"; "Same transformation across partitions" -> "Parallel Sweep"; "Codebase audit / assessment" -> "Multi-Dimensional Audit"; "Greenfield project kickoff" -> "Full Lifecycle"; } ``` | Strategy | When | Agents | Background | Key Pattern | | --------------------------- | ---------------------------------------- | --------- | ---------- | --------------------------------------------- | | **Research Swarm** | Knowledge gathering, docs, SOTA research | 10-60+ | Yes (100%) | Fan-out, each writes own doc | | **Epic Parallel Build** | Plan with independent epics/features | 20-60+ | Yes (90%+) | Wave dispatch by subsystem | | **Sequential Pipeline** | Dependent tasks, shared files | 3-15 | No (0%) | Implement -> Review -> Fix chain | | **Parallel Sweep** | Same fix/transform across modules | 4-10 | No (0%) | Partition by directory, fan-out | | **Multi-Dimensional Audit** | Quality gates, deep assessment | 6-9 | No (0%) | Same code, different review lenses | | **Full Lifecycle** | New project from scratch | All above | Mixed | Research -> Plan -> Build -> Review -> Harden | --- ## Strategy 1: Research Swarm Mass-deploy background agents to build a knowledge corpus. Each agent researches one topic and writes one markdown document. Zero dependencies between agents. ### When to Use - Kicking off a new project (need SOTA for all