GM-RKB Fleet
Jump to navigation
Jump to search
A GM-RKB Fleet is a fleet that consists of the AGET agents that collectively maintain and operate the GM-RKB knowledge base at www.gabormelli.com/RKB.
- AKA: GM-RKB AGET Fleet, GM-RKB Agent Portfolio, RKB Portfolio, GM-RKB Multi-Agent Fleet.
- Context:
- It can typically include a Supervisor AGET plus multiple domain-specialized Worker AGETs under a unified fleet governance policy.
- It can typically operate against a shared substrate comprising the GM-RKB MediaWiki instance hosted on Amazon EC2.
- It can typically depend on a web service stack combining Varnish, nginx, PHP-FPM, and MediaWiki for content delivery.
- It can typically coordinate its member agents via cross-fleet coordination mechanisms.
- It can typically exchange substrate-OK protocol messages across umbrella issues for fleet synchronization.
- It can typically share L-doc learning persistence and agent memory across its member agents.
- It can typically maintain fleet-level evidence chains that link individual agent actions to knowledge base updates.
- It can typically include member agents with distinct north stars but a shared fleet mission.
- It can typically enforce a common governance policy across all its member agents.
- It can typically orchestrate portfolio tasks through its Supervisor AGET.
- It can typically assign content enhancement tasks to specialized Worker AGETs.
- It can typically monitor substrate health through substrate diagnostics.
- It can typically remediate capacity issues via capacity remediation procedures.
- It can typically publish social media posts that are aware of GM-RKB wiki page content.
- It can often invoke large language model tools for concept page content generation.
- It can often perform wiki link enrichment across GM-RKB wiki pages.
- It can often validate concept page structure against house style rules.
- It can often escalate substrate incidents to a substrate-diagnostics worker agent.
- It can often batch knowledge base updates to reduce substrate load.
- It can often reconcile concurrent edits across member agents.
- It can often surface quality metrics for fleet monitoring.
- It can often adapt its task allocation based on agent availability.
- It can range from being a Single-Worker Fleet to being a Multi-Worker Fleet, depending on its fleet size.
- It can range from being a Read-Only Fleet to being a Read-Write Fleet, depending on its substrate permission.
- It can range from being a Manually-Triggered Fleet to being an Autonomously-Operating Fleet, depending on its automation level.
- It can range from being a Loosely-Coupled Fleet to being a Tightly-Coordinated Fleet, depending on its coordination intensity.
- It can range from being a Content-Focused Fleet to being an Infrastructure-Focused Fleet, depending on its operational scope.
- Examples:
- Worker AGETs, such as:
- A content-enhancement worker agent for content enhancement of GM-RKB wiki pages via large language model tools.
- A substrate-diagnostics worker agent for EC2 substrate diagnostics and capacity remediation on the GM-RKB substrate.
- A social-media worker agent for GM-RKB wiki page-aware social media post drafting and publication.
- Supervisor AGETs, such as:
- Coordination Mechanisms, such as:
- The substrate-OK protocol for cross-fleet coordination across umbrella issues.
- Worker AGETs, such as:
- Counter-Examples:
- A single standalone AGET agent, which lacks fleet-level coordination.
- A multi-agent system without shared AGET Framework governance and L-doc learning persistence.
- A monolithic wiki bot, which lacks member agent specialization.
- A human editorial team, which lacks AGET Framework agent automation.
- See: AGET Framework, Fleet (AGET), Worker AGET, Supervisor AGET, Cross-Fleet Coordination, GM-RKB, Multi-Agent System, L-doc, Substrate-OK Protocol, MediaWiki, Knowledge Base.
References
- The GM-RKB fleet operates the GM-RKB knowledge base published at www.gabormelli.com/RKB.
- AGET Framework concepts including Fleet (AGET), Worker AGET, and Supervisor AGET define the multi-agent system structure underlying this fleet.