mirror of
https://github.com/google-gemini/gemini-cli.git
synced 2026-03-12 07:01:09 -07:00
616062bdf6
This commit upgrades the agent with a robust self-validation workflow while ensuring 100% verbatim coverage of the original system prompt text. By moving to an additive model, we preserve all original reasoning anchors, instructional lead-ins, and senior engineering heuristics while injecting critical autonomous mandates. Verbatim Restoration: - All 'Context Efficiency' guidelines, lead-ins, and scenarios (Search/Understand/Navigate). - All 'Engineering Standards' regarding style mimicry, abstractions, and debt isolation. - Full 'Primary Workflows' sequence and formatting. Self-Validation Workflow Injections: - Research Phase: Parallel Discovery (manifests + logic) and High-Signal Grep. - Bug Fixing: Negative Verification (confirming repro failure) and Coverage Expansion. - Implementation: Transactional Edits (logical batching of module changes). - Validation Loop: Tiered Validation (Fixers -> Fast-Path -> Related Tests) and Smart Log Navigation (Tail-First). Technical Verification: - Verified against 67 core prompt tests and 14 snapshots. - New behavioral eval suite passed (evals/self_validation_workflow.eval.ts). - Full 'npm run preflight' successful.