Assemble context bundles for Planning, Implementation, and QA sub-agents.
npx skills add https://github.com/bketelsen/truenas-mcp --skill compose-team
Please help me install this Agent Skill. Skill: compose-team Source: https://github.com/bketelsen/truenas-mcp/tree/main/.gaai/core/skills/delivery/compose-team Command: npx skills add https://github.com/bketelsen/truenas-mcp --skill compose-team
This Skill ensures that each sub-agent receives precisely the context it needs to perform its task, preventing token waste and potential AI drift from over-contextualization, while also avoiding failures due to insufficient information.
Activate the compose-team skill after evaluate-story to prepare the context bundles for the next sub-agent.
High-intent search queries and answers about installing and using this skill.
To assemble context bundles for sub-agents and prevent AI drift, evaluate your story output first, then use this skill to resolve memory file paths and construct tailored, minimal context packages for each agent.
Managing context for planning, implementation, and QA agents requires constructing specific context packages from evaluated story output, ensuring each agent receives only the relevant files needed to optimize performance.
Dynamic memory file path resolution for agent coordination in a delivery pipeline is handled by reading an index to retrieve up-to-date memory categories, which ensures sub-agents receive accurate contextual information.
Yes, you can adjust context bundles for different agent tiers in microservices orchestration by applying defined rules that tailor the context package based on whether the tier is MicroDelivery, Core Team, or Specialists.
Sub-agents fail or drift during implementation tasks due to over-contextualization causing token waste or insufficient information failing the task, which is resolved by providing minimal, relevant context bundles.
Yes, you need to evaluate a story before assembling agent context bundles, because the evaluated story output and defined rules serve as the required input for resolving memory paths and constructing the packages.
petrSimonidesXartAssemble per-sub-agent context bundles from evaluate-story output for deterministic spawning.
RobinNorbergAutomate end-to-end software delivery from product idea to working code.
ricableCoordinate distributed AI agents with shared vector memory and task routing.
OpenAnalystIncAnalyze token usage and compress prompts in long-running agent sessions.
