# Phase 2: Tier 1-2 Providers - Context **Gathered:** 2026-04-05 **Status:** Ready for planning **Mode:** Auto-generated (infrastructure phase — discuss skipped) ## Phase Boundary The 26 highest-value LLM provider YAML definitions exist with accurate regex patterns, keyword lists, confidence levels, and verify endpoints — covering OpenAI, Anthropic, Google AI, AWS Bedrock, Azure OpenAI and all major inference platforms. ## Implementation Decisions ### Claude's Discretion All implementation choices are at Claude's discretion — pure infrastructure/data phase. Use ROADMAP phase goal, success criteria, and existing provider YAML schema (providers/openai.yaml, providers/anthropic.yaml, providers/huggingface.yaml) as templates. ## Existing Code Insights ### Reusable Assets - providers/openai.yaml, providers/anthropic.yaml, providers/huggingface.yaml — reference YAML definitions - pkg/providers/schema.go — Provider struct with Name, Keywords, Patterns, Tier, Category, VerifySpec - pkg/providers/loader.go — embed.FS YAML loader - pkg/providers/registry.go — Registry with AC automaton built from all keywords ### Established Patterns - YAML format: name, display_name, category, tier, format_version, last_verified, keywords[], patterns[], verify - Patterns have regex, entropy_min, confidence fields - Keywords are lowercase terms used for Aho-Corasick pre-filtering ### Integration Points - providers/ directory — embed.FS reads all *.yaml files at compile time - Registry.AC() rebuilds automatically with new provider keywords ## Specific Ideas No specific requirements — infrastructure phase. Refer to ROADMAP phase description and success criteria. ## Deferred Ideas None — discuss phase skipped.