Lobster Coordinator

Production-grade 3-layer agent orchestration with dual-blind verification and automated model routing.

Install
cmdop skills install agensi-lobster-coordinator

Lobster Coordinator is a skill that implements a three-layer architecture for orchestrating multi-agent AI workflows. The three layers are AgentTool, which handles task delegation to individual agents; Swarm, which manages team-level messaging between agents; and Coordinator, which handles global task decomposition and result aggregation across the entire system.

The skill addresses a specific failure mode in multi-agent systems: agent loops, where agents get stuck in repetitive cycles without making progress. It tackles this through secondary verification layers and strict multi-stage review processes drawn from the Santa Method and Superpowers frameworks.

Code review follows a two-stage process. The first stage checks for specification compliance; the second checks code quality. The Santa Method adds a dual-blind verification step requiring two independent reviewers to approve code before it is merged. This reduces the risk of hallucinated outputs and accumulated technical debt in long-running agentic pipelines.

Model routing allows tasks to be assigned to different model tiers — Cheap, Standard, or Elite — based on their complexity, which can reduce inference costs. Agents report status as DONE, BLOCKED, or NEED_CONTEXT, giving the coordinator structured state information to manage task progression.

This skill is built specifically for advanced AI environments such as OpenClaw, Claude Code, and Cursor, and relies on capabilities like sessions_spawn for isolated sub-agent execution. It is not a general-purpose MCP server and has no database or API integration tools.

Use cases

  • Orchestrate parallel sub-agent pipelines where tasks must be decomposed, delegated, and aggregated
  • Enforce automated two-stage code review (spec compliance then code quality) in AI-generated codebases
  • Apply dual-blind Santa Method verification before merging AI-produced code
  • Route agent tasks to cost-appropriate model tiers based on task complexity
  • Manage BLOCKED or NEED_CONTEXT agent states to prevent pipeline stalls
  • Coordinate multi-agent workflows inside Claude Code, OpenClaw, or Cursor environments

When to use it

  • Building multi-agent systems where agent loop failures are a known risk
  • Running AI coding pipelines that require structured, multi-stage review before code is accepted
  • Operating inside Claude Code, OpenClaw, or Cursor where sessions_spawn is available
  • When cost control through model-tier routing is a priority across complex agentic tasks

When not to use it

  • Connecting to Postgres, databases, or any external APIs — this skill has no such tools
  • Single-agent workflows that do not require delegation or coordination layers
  • Environments that do not support sessions_spawn or equivalent sub-agent spawning capabilities
  • Projects requiring an MCP server with defined tool endpoints — this skill exposes no tools
  • Non-agentic scripting or automation tasks outside AI orchestration contexts