Module model_tier 

Module: model_tier

Spec

• Defines Tier: a harness-agnostic complexity bucket (Auto, Low, Med, High) used to classify task complexity and gate model selection.
• Tier derives PartialOrd such that Auto < Low < Med < High. Gating is a simple > comparison: a task whose effective tier exceeds the running model’s tier should prompt the user to switch.
• Defines ModelConfig (under [model] in the global TOML config) and TierMap (per-harness tier→model name resolution under [model.tiers.<harness>]).
• detect_harness() reads environment variables (CLAUDE_CODE_SESSION, CODEX_SESSION) to identify the active agent harness, falling back to "default".
• resolve_tier_to_model(tier, harness, config) maps a Tier to the concrete model name configured for the given harness, falling back to built-in defaults for the claude-code, codex, and default harnesses.
• tier_from_model_name(name, harness, config) is the reverse lookup: given a concrete model name (e.g., "opus"), find the tier it belongs to in the harness mapping.
• Tier::FromStr accepts case-insensitive auto | low | med | high.

Agentic Contracts

• Total ordering: Tier implements PartialOrd deterministically; gating logic is a single comparison and can be safely executed by any model tier.
• Auto is the lowest: Tier::Auto represents “no preference” and compares less than Low. The effective_tier composition treats Auto as “fall through to next source.”
• Built-in defaults: when no [model.tiers.<harness>] section is present, the resolver falls back to compiled-in maps for known harnesses. This means a fresh install needs zero config for the common case.
• Reverse lookup is partial: tier_from_model_name returns None if the model name doesn’t appear in any tier slot for the harness. Callers should treat None as “unknown — leave tier as Auto.”

Evals

• tier_ordering: Auto < Low < Med < High holds for <, >, <=, >=.
• tier_from_str_case_insensitive: "LOW", "low", "Low" all parse to Tier::Low.
• tier_from_str_invalid: unknown strings return Err.
• harness_detection_default: with no env vars set, detect_harness() returns "default".
• resolve_builtin_claude_code: resolve_tier_to_model(Tier::High, "claude-code", &Config::default()) returns Some("opus").
• resolve_unknown_harness_uses_default: an unknown harness falls through to the "default" built-in map.
• tier_from_model_name_roundtrip: tier_from_model_name("opus", "claude-code", ...) returns Some(Tier::High).

Structs

ModelConfig

Global [model] config section.

ModelSwitchScan

Result of scanning a unified diff for an inline /model <x> command.

TierMap

Per-harness tier → concrete model name map.

Enums

Tier

Harness-agnostic model complexity tier.

Functions

component_value_to_tier

Resolve a <!-- agent:model --> component value to a Tier.

compose_effective_tier

Compose the final effective_tier from all available sources.

detect_harness

Detect the active harness from environment variables.

extract_model_component

Extract the value inside a <!-- agent:model -->...<!-- /agent:model --> component.

parse_model_arg

Parse a /model <arg> argument: either a tier name (low|med|high) or a concrete model name (opus|sonnet|...).

resolve_tier_to_model

Resolve a Tier to a concrete model name for the given harness.

scan_model_switch

Scan a unified diff for an inline /model <x> command in user-added lines.

suggested_tier

Compute a suggested_tier from structural diff signals.

tier_from_model_name

Reverse lookup: given a concrete model name, find its tier in the harness’s mapping.