Red Hat's MIT ai-code-review already implements phases 1-2 (four forge clients,
six providers, CI integration, repo context file). Adds a research writeup,
inserts Phase 0 (evaluate it before building), and folds in seven requirements
the original design missed — chiefly prior-comment synthesis, without which our
own 1.7-runs-per-PR assumption means every push re-posts dismissed findings.
Amends implementation tasks 4, 5, 6, 9, 10 and gates the subagent briefs.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_011Ye1KNFMkkUtmzTypHXkoK
Per-task briefs a cold subagent can execute without conversation history:
objective, deliverables, constraints, done criteria, likely failure modes.
Adds orchestrator instructions (dependency graph, model tiers, dispatch loop,
stopping conditions), a shared context block, a review gate checklist, and
notes for running the plan on a non-Claude model.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_011Ye1KNFMkkUtmzTypHXkoK
TDD task breakdown for the walking skeleton: diff parsing, forge adapter port
with a local git implementation, rules-based tiering, config loading, analyzer
runner behind a ModelClient port, pi SDK client, aggregator, JSONL emitter, CLI.
Later phases recorded as milestones with done-when criteria.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_011Ye1KNFMkkUtmzTypHXkoK
Marcos