Files
reliquary/plugins/spire/skills/igor/references/model-tiers.md
T
g4borg e054ffd2bd ✨ Spire plugin — MVP build (phases 1-3)
Complete plugin at plugins/spire/ replacing g4b_ai:
- Igor (session coordinator skill + agent)
- Adept (domain knowledge agent)
- Athanor (dispatching orchestrator, "the oven")
- Tome (generic knowledge layer, opt-in)
- Ported from g4b_ai: workpad, research-tree, research,
  implementation, svg skills + researcher/search/scout agents

All warpspace-specific content stripped. Tome genericized.
Project-specific config via shadowed project.md reference.
2026-09-14 14:28:26 +02:00

2.4 KiB

Model Tiers for Delegation

Prefer a project-pinned agent from .claude/agents/ when one matches the task; the tiers below are the fallback and the rationale.

Task type Tier Rationale
Domain knowledge, repeated consultation adept agent (Opus, long-lived) Accumulated priors, informed search, journal continuity; domain-scoped
Deep research, bug investigation Opus-tier research agent Depth, reasoning; writes reports, never modifies code
Planned implementation implementer agent Executes plans, verifies with the project's check/test commands
Multi-part jobs needing their own subagent fan-out athanor (Opus-tier, may use Agent/Task tool) Decomposes a brief and runs its own subagents
Complex summaries, git fixes, small code changes Sonnet subagent Capability/speed balance
Test runs, compiles, commits, linear-data summaries Haiku subagent Fast, cheap, mechanical

Rules of thumb:

  • One task -> one subagent with an explicit output file path in the workpad.
  • A job with 3+ independent subtasks, or research + implementation combined, goes to athanor as a single brief (scope, output paths, completion criteria).
  • When unsure between tiers, prefer the cheaper tier and escalate on failure.

Sub-delegation guard:

  • Sub-delegation guard. Agent types without the Agent tool cannot spawn children: Explore (read-only search) and Plan (design) are safe leaves. general-purpose and research/implementer agents may have the Agent tool and can sub-delegate — give them explicit "do not delegate" instructions, or route fan-out work to athanor on purpose. Only ever expect a second, differently-task-id'd notification when you dispatched something that was allowed to fan out.

Model selection rules:

  • Never dispatch a Fable-tier subagent unless the user explicitly asks. Always pass model explicitly on Agent calls ("opus", "sonnet", "haiku") — omitting it inherits the coordinator's model, which may be Fable. Subagents are Opus or lower.
  • Prefer Opus 4.6 for implementation and detail-precise lookups; Opus 5 only for reviews. Pinned implementer/researcher agents should be Opus 4.6. subagent_type: "general-purpose" + model: "opus" resolves to Opus 5, which tends to hallucinate — use it for review/critique-style tasks only.

This table can grow — add new agents and heuristics here as they prove out.