You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
Allow setting different models for different squad members #205
I'd like to be able to specify the models that my different squad members use eriod in some cases I'd like to specify that they use the inexpensive or free GPT models or I might want to specify that they can use the latest and greatest most expensive models.
Can we introduce a standard for a way to specify these perhaps in a Agents charter or be able to provide a command line that allows us to blanket specify a model that's going to be used by any or all team members
Thanks for raising this! Model configuration and agent customization is something we're actively working on. We've made progress on related areas like charter-based governance (see #242 and our recent architecture work), and we're building out the foundations that'll support per-agent model selection. If you're still looking to test this approach, feel free to reopen or file a follow-up issue with your specific use case — we'd love to hear how you're thinking about using different models. 🙌
Thanks for pushing on model flexibility, @csharpfritz! Being able to specify which models different squad members use—cheaper for routine work, premium for complex reasoning—is a huge UX gap. We've got this in active design now. Charter-level configuration is the right mental model. Really appreciate this input! 🚀
Activity
feat: CI/CD workflows + squad resolution algorithms
docs(ai-team): Merge next-wave triage decisions, update agent history
fix: wire missing CLI commands into cli-entry.ts (#244)
github-actions commentedon Mar 7, 2026
📋 Assigned to Keaton (Lead)
Issue: #205 — Allow setting different models for different squad members
Keaton will pick this up in the next Copilot session.
github-actions commentedon Mar 7, 2026
🏗️ Squad Triage — Keaton (Lead)
Issue: #205 — Allow setting different models for different squad members
Assigned to: Hockney (Tester)
Reason: Issue relates to testing/quality work
@copilot evaluation: No strong capability match — routed to squad member.
Team roster:
squad:keatonsquad:verbalsquad:fenstersquad:hockneysquad:mcmanussquad:kujansquad:ediesquad:kobayashisquad:fortiersquad:rabinsquad:baersquad:redfootsquad:strauszsquad:saulsquad:kovashsquad:marquezsquad:cherittosquad:breedansquad:natesquad:waingrosquad:ralphsquad:copilotbradygaster commentedon Mar 7, 2026
Thanks for raising this! Model configuration and agent customization is something we're actively working on. We've made progress on related areas like charter-based governance (see #242 and our recent architecture work), and we're building out the foundations that'll support per-agent model selection. If you're still looking to test this approach, feel free to reopen or file a follow-up issue with your specific use case — we'd love to hear how you're thinking about using different models. 🙌
bradygaster commentedon Mar 7, 2026
Thanks for pushing on model flexibility, @csharpfritz! Being able to specify which models different squad members use—cheaper for routine work, premium for complex reasoning—is a huge UX gap. We've got this in active design now. Charter-level configuration is the right mental model. Really appreciate this input! 🚀
— McManus 📝, on behalf of the Squad